CRC32 of checkout's path is now part of dev-container's instance
name.
TEST=`docker ps` shows crc of script's absolute path in checkout
FIXED=b:237592974
Change-Id: I7179c40e5c8436a5866d5d75b01696dc788725c9
Reviewed-on: https://chromium-review.googlesource.com/c/crosvm/crosvm/+/3790038
Reviewed-by: Dennis Kempin <denniskempin@google.com>
Commit-Queue: Dennis Kempin <denniskempin@google.com>
Tested-by: Zihan Chen <zihanchen@google.com>
Auto-Submit: Zihan Chen <zihanchen@google.com>
The ChromeOS toolchain is updated to this point, so bring our crosvm
testing in line.
BUG=b:239075544
TEST=tools/presubmit --all
Change-Id: I6bfa7bfd802f1cd75204ba59b560de176a56326f
Reviewed-on: https://chromium-review.googlesource.com/c/crosvm/crosvm/+/3764424
Tested-by: Daniel Verkamp <dverkamp@chromium.org>
Reviewed-by: Alexandre Courbot <acourbot@chromium.org>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
crosvm is switching the import style to use one import per line.
While more verbose, this will greatly reduce the occurence of merge
conflicts going forward.
Note: This is using a nightly feature of rustfmt. So it's a one-off
re-format only. We are considering adding a nightly toolchain to
enable the feature permanently.
BUG=b:239937122
TEST=CQ
Change-Id: Id2dd4dbdc0adfc4f8f3dd1d09da1daafa2a39992
Reviewed-on: https://chromium-review.googlesource.com/c/crosvm/crosvm/+/3784345
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: Dennis Kempin <denniskempin@google.com>
Commit-Queue: Dennis Kempin <denniskempin@google.com>
The test_target.py script is not longer directly executable
but the test runner would still be pointed at it when used
for iterative development.
BUG=None
TEST=./tools/test_target set vm:aarch64 && source .envrc && cargo test
Change-Id: Id179fe83cd1f449d266576340a2f6f46e63c2878
Reviewed-on: https://chromium-review.googlesource.com/c/crosvm/crosvm/+/3784340
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: Dennis Kempin <denniskempin@google.com>
Commit-Queue: Dennis Kempin <denniskempin@google.com>
When running tools/cl on a branch without an upstream set, ask if the
user would like the script to fix it automatically.
BUG=None
TEST=tools/cl upload # on a new branch with no upstream
Change-Id: Id5f4ee30058a209ac352013159d09278cd92a23b
Reviewed-on: https://chromium-review.googlesource.com/c/crosvm/crosvm/+/3780873
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
Reviewed-by: Dennis Kempin <denniskempin@google.com>
Commit-Queue: Dennis Kempin <denniskempin@google.com>
Auto-Submit: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: Daniel Verkamp <dverkamp@chromium.org>
It was missing in some commands. Also runs them when no command
is specified. This will ensure the git hook is installed and
print documentation when calling: "./tools/cl"
BUG=b:239823229
TEST=./tools/cl
Change-Id: Id2248310011f4692cf477ac51d52b578d2f00a8d
Reviewed-on: https://chromium-review.googlesource.com/c/crosvm/crosvm/+/3780868
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Auto-Submit: Dennis Kempin <denniskempin@google.com>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
Commit-Queue: Dennis Kempin <denniskempin@google.com>
Tested-by: Dennis Kempin <denniskempin@google.com>
Exclude crash-report feature when generating cargo-doc because the
feature only works on Windows. This should fix GitHub action's failure
started by crrev.com/c/3765346 [1].
[1] https://github.com/google/crosvm/runs/7417889096?check_suite_focus=true
BUG=none
TEST=run ./tools/cargo-doc locally
Change-Id: I1481bf8221fc7a583b9147992d116278b4df9e05
Reviewed-on: https://chromium-review.googlesource.com/c/crosvm/crosvm/+/3780081
Auto-Submit: Keiichi Watanabe <keiichiw@chromium.org>
Commit-Queue: Vikram Auradkar <auradkar@google.com>
Reviewed-by: Vikram Auradkar <auradkar@google.com>
Tested-by: Keiichi Watanabe <keiichiw@chromium.org>
Mount the cros root instead so the git worktree or symlink can be
resolved.
BUG=b:234322274
TEST=./tools/dev_container
Change-Id: I52d0cbe5e9a38cb6febcf1f328970e13209bcf9f
Reviewed-on: https://chromium-review.googlesource.com/c/crosvm/crosvm/+/3768048
Tested-by: Dennis Kempin <denniskempin@google.com>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Commit-Queue: Dennis Kempin <denniskempin@google.com>
Auto-Submit: Dennis Kempin <denniskempin@google.com>
This cleans up the output a bit when there are many local branches that
don't have an upstream configured.
BUG=None
TEST=tools/cl prune
Change-Id: Ic5892dcbca315613682326c5615fce28d49b421d
Reviewed-on: https://chromium-review.googlesource.com/c/crosvm/crosvm/+/3764046
Reviewed-by: Dennis Kempin <denniskempin@google.com>
Tested-by: Daniel Verkamp <dverkamp@chromium.org>
Commit-Queue: Dennis Kempin <denniskempin@google.com>
Auto-Submit: Daniel Verkamp <dverkamp@chromium.org>
The prune command will delete all branches with gerrit changes
that have been submitted.
BUG=None
TEST=./tools/cl prune
Change-Id: I2f942591e6e29a16d1ed6655655ef9f8cb4fd34f
Reviewed-on: https://chromium-review.googlesource.com/c/crosvm/crosvm/+/3765345
Commit-Queue: Dennis Kempin <denniskempin@google.com>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: Dennis Kempin <denniskempin@google.com>
These will be useful in other scripts as well.
BUG=None
TEST=./tools/chromeos/mergebot update-dry-runs
Change-Id: Id75ed458fdd212d4564a547274209b1f39796ceb
Reviewed-on: https://chromium-review.googlesource.com/c/crosvm/crosvm/+/3765344
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Commit-Queue: Dennis Kempin <denniskempin@google.com>
Tested-by: Dennis Kempin <denniskempin@google.com>
This was hard-coded to argh, which is the only package that uses it for
now, but the function may be used elsewhere in the future.
Also don't print a newline so the user can answer the prompt on the same
line after the [y/N].
BUG=None
TEST=Run tools/cl without python3-argh installed
Change-Id: I587fd1ded81f21724169a2fe90445f6055d1f421
Reviewed-on: https://chromium-review.googlesource.com/c/crosvm/crosvm/+/3764375
Tested-by: Daniel Verkamp <dverkamp@chromium.org>
Reviewed-by: Dennis Kempin <denniskempin@google.com>
Auto-Submit: Daniel Verkamp <dverkamp@chromium.org>
Commit-Queue: Dennis Kempin <denniskempin@google.com>
Ask the user if they want to 'pip install' the dependency
if it is missing.
Not the prettiest solution, but helps make tools like
./tools/cl usable with minimal dependencies available.
Eventually we may consider pyenv or a similar solution.
BUG=b:239072471
TEST=pip uninstall argh; ./tools/cl
Change-Id: I4872e86883a9a0903a408d2f6d4f35dd1841f2ef
Reviewed-on: https://chromium-review.googlesource.com/c/crosvm/crosvm/+/3764425
Auto-Submit: Dennis Kempin <denniskempin@google.com>
Commit-Queue: Dennis Kempin <denniskempin@google.com>
Tested-by: Dennis Kempin <denniskempin@google.com>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
BUG=None
TEST=tools/presubmit --all # with a failed test
Change-Id: I67262353cbd732b1394cdbe60b38d04c3a6e817b
Reviewed-on: https://chromium-review.googlesource.com/c/crosvm/crosvm/+/3763359
Reviewed-by: Dennis Kempin <denniskempin@google.com>
Tested-by: Dennis Kempin <denniskempin@google.com>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
This updates all references to chromiumos/platform/crosvm to
crosvm/crosvm.
It will also switch ./tools/cl to upload to the new main repository.
BUG=None
TEST=None
Change-Id: I1f71e5d5de2300e57ace1315cb4553f68533d576
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3760903
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: Dennis Kempin <denniskempin@google.com>
The example is currently building a crosvm with gpu support, which
doesn't link properly out of the box. Now, we build explicitly with all
features off, which is sufficient to run the example.
BUG=none
TEST=ran the example.
Change-Id: I1699741f9c094831827b090782be5f5d7492eefa
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3756725
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Noah Gold <nkgold@google.com>
This shuts off some of the Windows serial code beacuse it's blocked on a
separate CL that upstreams bits of the serial device. (It doesn't work
in this repo yet anyway.)
BUG=b:237597358
TEST=builds
Change-Id: If6b223ff29d55f44bff7d006c1e0b5db3733a4ac
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3741429
Reviewed-by: Vikram Auradkar <auradkar@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Noah Gold <nkgold@google.com>
Some tests can fail if there is interference from other tests (e.g.
EINTR errors).
BUG=b:238232551
TEST=./tools/presubmit
Change-Id: I962a030771cee9d77f309d83fb7f0f43c18ada45
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3760897
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Always test the script, even when just adding a comment...
BUG=None
TEST=./tools/install-doc-deps
Change-Id: I713d45f7dbfe7c6b7fc319399173749e9c6dce84
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3758181
Commit-Queue: Alexandre Courbot <acourbot@chromium.org>
Reviewed-by: Keiichi Watanabe <keiichiw@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
This is a reland of commit 213f9fe8a7.
In light of the upcoming VAAPI video decoder backend, add a VAAPI
wrapper crate that exposes a safe Rust API for a subset of the VAAPI C
code. This crate will be called from the VAAPI video decoder backend in
order to decode frames.
BUG=b:214478588
TEST=cargo build --features "video-decoder,vaapi"
TEST=`cargo test -- --include-ignored` in `media/libva` passes on a
device with Intel GPU and libva installed.
Change-Id: I586a160e477e466985c5cfa65a527542ddc03226
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3752274
Reviewed-by: Keiichi Watanabe <keiichiw@chromium.org>
Commit-Queue: Alexandre Courbot <acourbot@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
libva introduced a dependency on libudev, and the latter does not
distinguish between regular builds and runs of cargo doc to avoid
depending on the system library. Thus add it as a dependency to be able
to successfully build the docs.
BUG=b:214478588
TEST=./tools/cargo-doc
Change-Id: I4150b312fd7e1a5197c817e4d7cee5766a3c569a
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3758165
Reviewed-by: Keiichi Watanabe <keiichiw@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Alexandre Courbot <acourbot@chromium.org>
This reverts commit 213f9fe8a7.
Reason for revert: breaks the ChromeOS builder.
Original change's description:
> Add a VAAPI wrapper crate
>
> In light of the upcoming VAAPI video decoder backend, add a VAAPI
> wrapper crate that exposes a safe Rust API for a subset of the VAAPI C
> code. This crate will be called from the VAAPI video decoder backend in
> order to decode frames.
>
> BUG=b:214478588
> TEST=cargo build --features "video-decoder,vaapi"
> TEST=`cargo test -- --include-ignored` in `media/libva` passes on a
> device with Intel GPU and libva installed.
>
> Change-Id: I4afa96c49d045251827b97bd78faeba57575aedc
Bug: b:214478588
Change-Id: Ib5a88cd4c5fdd2df2e69fd3a0896ee789585840d
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3752267
Auto-Submit: Alexandre Courbot <acourbot@chromium.org>
Commit-Queue: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Reviewed-by: Keiichi Watanabe <keiichiw@chromium.org>
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
In light of the upcoming VAAPI video decoder backend, add a VAAPI
wrapper crate that exposes a safe Rust API for a subset of the VAAPI C
code. This crate will be called from the VAAPI video decoder backend in
order to decode frames.
BUG=b:214478588
TEST=cargo build --features "video-decoder,vaapi"
TEST=`cargo test -- --include-ignored` in `media/libva` passes on a
device with Intel GPU and libva installed.
Change-Id: I4afa96c49d045251827b97bd78faeba57575aedc
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3422779
Reviewed-by: Alexandre Courbot <acourbot@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Alexandre Courbot <acourbot@chromium.org>
Reviewed-by: Keiichi Watanabe <keiichiw@chromium.org>
Normally we run test binaries in parallel, this change allows some to be
executed exclusively.
This fixes some of the odd issues and flakes we have been seeing.
integration_tests can hang if run in parallel with other tests and
cros_async has the odd habit of throwing the test_runner process into
the background.
Neither happens when the tests are the only ones being executed.
Also adds the capability to specify which test binaries to run,
which was helpful in debugging this issue.
BUG=b:238232551
TEST=./tools/run_tests --repeat 10
Change-Id: Ie5b82e329007399e95bf38111ec6e7cb6761ec0c
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3751027
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
The test runner can now repeat tests in two ways:
- Via --repeat to run tests multiple times and fail if any one of them
fails. Which will help us find flakes in post-submit.
- Via --retry, which will retry a test if it has failed. Which makes
pre-submit tests more resilient to flakes.
Both can be configured by builder configs to adjust as needed.
Also slighly changes the repeat behavior to repeat tests in batches, so
we do not run the same test multiple times simultaneously, which can
cause some of them to fail. It's also easier to read the results.
BUG=b:238232551
TEST=added a random flake into some tests and used --repeat and --retry
Change-Id: I78e0ff0751da40a99a56080d9baf207307c9e93a
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3751835
Tested-by: Dennis Kempin <denniskempin@google.com>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
This Allows Luci builders to cache cargo home between builds. So we
do not have to download so many third party crates with each build.
CARGO_HOME is specifically intended to be cached in CI systems.
BUG=b:233230027
TEST=CROSVM_CONTAINER_CACHE=/tmp/test ./tools/dev_container --clean
cargo build
Change-Id: I11580c5ed3151519ece4a651cb22d059c7c3eb87
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3739368
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Dennis Kempin <denniskempin@google.com>
The format string wasn't actually formatted, so it just printed out the
literal "{shorthand}" string in the error message.
BUG=None
TEST=tools/run_tests --target=host --build-target=bogus
Change-Id: Idad47064db3112a37df3a630b92a4a4da8da82de
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3750069
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
Reviewed-by: Dennis Kempin <denniskempin@google.com>
CrOS checkouts are experimenting with using git worktree, where
the path to the hook directory is not .git/hooks. So ask git
where the hooks are stored instead.
BUG=b:236645893
TEST=./tools/cl upload in a worktree checkout
Change-Id: Ib353da0aea5d27ce29d1c082268a326adb429d23
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3739373
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Junichi Uekawa <uekawa@chromium.org>
Commit-Queue: Dennis Kempin <denniskempin@google.com>
Updates run_tests to use cargo style target triples for specifying
build targets. A simple 'aarch64' or 'armhf' was nice while we just
had linux builds. We now are looking at windows and possibly
different toolchain options (e.g. msvc vs gnu), so our old system
was getting confusing and inconsistent.
We used to have some special handling for adding wrappers to test
runs for emulation (e.g. wine, qemu). That logic has been moved
into TestTarget which now contains not just where to run the test
but also how.
Supported are armhf/aarch64 qemu as well as wine64.
The CLI has been updated to match and now uses the build-target
argument instead of arch.
The following combinations have been tested (though not all
combinations actually pass all tests, which is a separate issue).
./tools/run_tests
./tools/run_tests --target=host --build-target=x86_64-unknown-linux-gnu
./tools/run_tests --target=host --build-target=armhf
./tools/run_tests --target=host --build-target=aarch64
./tools/run_tests --target=host --build-target=mingw64
./tools/run_tests --target=vm:aarch64
./tools/run_tests --target=vm:aarch64 --build-target=armhf
BUG=b:233914170
TEST=See above
Change-Id: Ic6dbb5b39788e2573714606d3bb0e7c712032d91
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3739240
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Dennis Kempin <denniskempin@google.com>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Refactor udmabuf support out from virtio_gpu into the shared vm_memory
target, so that can be used for vvu.
BUG=b:201745804
TEST=compiles
TEST=./tools/bindgen-all-the-things
Change-Id: I2944f0cb8c75e4320400bfb9f7026dc1efe0c12b
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3716339
Reviewed-by: Keiichi Watanabe <keiichiw@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: David Stevens <stevensd@chromium.org>
Reviewed-by: Alexandre Courbot <acourbot@chromium.org>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
One more script that needs to run on python 3.8 so we can run test
for windows on luci.
BUG=b:234173142
TEST=vpython3 tools/run_tests
Change-Id: I0348a567b2edec3b5e3fca77f316f39d1e924adc
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3723037
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Use an INT3 (0xCC) instruction to cause the example to exit after
printing the message. This is more convenient than having to manually
kill crosvm from another terminal.
BUG=None
TEST=Run baremetal and observe that it exits
Change-Id: I4baeecca41d156c82bb1e1b27d0f8c2ba93959f9
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3691966
Reviewed-by: Anton Romanov <romanton@google.com>
Reviewed-by: Alexandre Courbot <acourbot@chromium.org>
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
Fix loading of the example baremetal kernel with the fixed ELF kernel
loader:
- Remove the "RAM" address space so virtual and physical address match.
- Remove the 0x200-byte padding now that entry point address is used.
BUG=b:234155022
TEST=Run baremetal as in tools/examples/baremetal/README.md
Change-Id: I61394cdf4bf71f91736da5a636b0088ecfe78c84
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3691965
Commit-Queue: Daniel Verkamp <dverkamp@chromium.org>
Reviewed-by: Anton Romanov <romanton@google.com>
Reviewed-by: Alexandre Courbot <acourbot@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
These libraries are needed to build the upcoming vaapi backend. Also
update the container so they are included.
BUG=b:214478588
TEST=./tools/dev_container ./tools/presubmit --all
Change-Id: Id30c38557d56b4c94ca5f2419c73b34520916ed0
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3715087
Reviewed-by: Dennis Kempin <denniskempin@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Alexandre Courbot <acourbot@chromium.org>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
rwroot automatically adds root=/dev/vd[a-z], but the example image
actually uses /dev/vda5. This leads to confusion when two root= kernel
command line arguments are passed in and only one of them works.
Therefore we use rwdisk instead which doesn't add the kernel argument.
BUG=b:236198584
TEST=mdbook builds, examples build and run
Change-Id: I9bfc05445b1e2aef197a43ee647d7c784d50dad7
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3708365
Commit-Queue: Yuanchu Xie <yuanchu@google.com>
Reviewed-by: Keiichi Watanabe <keiichiw@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
To enable the merge bot on luci we had to solve a couple of problems:
- We cannot use http cookies for auth, so added gcloud auth into
merge_bot.
- Switch back to original HEAD after running merge bot. Otherwise
the version of the merge_bot script can change between invocations.
- Do not soft reset when checking out crosvm source.
- For less invasive testing, also added a few more exceptions when
using MERGE_BOT_TEST so it won't email, etc.
- Rename to update_chromeos_merges, as it is more fitting to what
the bot is doing
BUG=b:233913643
TEST=https://ci.chromium.org/swarming/task/5b58ed4497fda510
Change-Id: Iaa9b4821b70a1e90d19637d01856196bd7852ed5
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3692686
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Tested-by: kokoro <noreply+kokoro@google.com>
The virtio video decoder device is currently only available under very
drastic conditions: a build linked against libvda (a ChromeOS-only
library that needs the cros chroot to be built and linked against), and
a ChromeOS-flavored Chrome instance running alongside crosvm, so the
browser can provide the video decoding service through Mojo.
This makes the decoder device very difficult to develop on for
non-Chromies, and also for Chromies actually since they will always need
a DUT to test it on.
This patch introduces an alternative decoder backend based on
ffmpeg's libraries that performs decoding on the host's CPU. It supports both
guest pages and virtio objects as target, and can be considered a
reliable and predictable way to test the decoder in any environment.
We introduce our own ffmpeg bindings after a quick state of the art
revealed that the existing ones were all unsuitable, either for
technical or licensing reasons. Doing so is also not a big effort and
does not add any new external crate dependency to crosvm.
BUG=b:169295147
TEST=cargo test --features "video-decoder,ffmpeg" -p devices ffmpeg
TEST=v4l2r's simple_decoder example decodes test-25fps.h264 properly with the
following command:
./simple_decoder test-25fps.h264 /dev/video0 --input_format h264 --save test-25fps.nv12
TEST=ARCVM Android youtube plays videos correctly when the ffmpeg
backend is used.
Change-Id: Ic9c586193f7939f2a3fe59d009c3666585a8bbc7
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3026355
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Alexandre Courbot <acourbot@chromium.org>
Reviewed-by: Keiichi Watanabe <keiichiw@chromium.org>
Add the ffmpeg libraries needed by the ffmpeg decoder device backend and
bump the dev_container and testvm to include them.
BUG=b:169295147
TEST=./tools/install-deps
TEST=./tools/dev_container
Change-Id: Ifc07b9599403aa1ed18a96067ada1fa2efa6e13c
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/crosvm/+/3685682
Reviewed-by: Junichi Uekawa <uekawa@chromium.org>
Reviewed-by: Dennis Kempin <denniskempin@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Commit-Queue: Alexandre Courbot <acourbot@chromium.org>
