To optimize for offloading success, the list of output layers can be
ordered by allocation priority, improving the chances of a higher
allocation score before the deadline.
The allocation algorithm picks DRM planes first, then tests output
layers against them. To reduce the number of tests, the layers can be
ordered such that it matches the DRM plane order. DRM planes in
libliftoff are ordered by primary, followed by all other planes in
descending z-order. We order the layers similarly, with the composition
layer first, followed by layers of descending priority.
In addition to layer priority, it's z-pos and intersection with other
layers are considered. Since to offload a high priority layer, all
intersection layers with higher z-pos will also need to be offloaded.
This also changes when reallocation is necessary. On top of the existing
criteria, reallocation is done when a layer's ordering changes.
With layer priority now considered, the priority test now passes --
enable it.
By default, drmModeGetFB2() behaves as-if the kernel didn't support
the IOCTL. That way we don't need to update all tests to set sensible
FB info.
If a test is specifically interested in testing drmModeGetFB2()
behavior, it can call liftoff_mock_drm_set_fb_info().
This was fine for simple test cases, but would need to become
more complicated for cases where e.g. zero FB_ID or alpha is
involved. Let's just hardcode the expected value in the test data
instead.
This allows to mimick i915's requirement of failing all commits
which don't have the primary plane enabled.
We could expand our test matrix to test with the param *and* without
it, but since enabling the param just makes the atomic check more
strict there's no real benefit of doing that. Enabling the param
can only make tests fail more often, not the other way around.
Right now, compositors are expected to query liftoff_layer_get_plane_id to
decide whether a layer needs to be composited.
However there are cases where a layer has no plane, but still doesn't need
composition (e.g. when FB_ID is zero or when alpha is zero, in the future when
the layer is out of CRTC bounds etc).
Add a new API to make users' life simpler.
Closes: https://github.com/emersion/libliftoff/issues/54
When the caller wants to perform a modeset, they will need to pass
DRM_MODE_ATOMIC_ALLOW_MODESET in the atomic commit flags, otherwise the driver
will fail the commit. libliftoff also needs to provide these flags in its
test-only commits.
In file included from ../log.c:4:
In file included from ../include/log.h:4:
In file included from ../include/libliftoff.h:8:
/usr/local/include/xf86drmMode.h:551:19: error: zero size arrays are an extension [-Werror,-Wzero-length-array]
uint32_t lessees[0];
^
/usr/local/include/xf86drmMode.h:558:19: error: zero size arrays are an extension [-Werror,-Wzero-length-array]
uint32_t objects[0];
^
