DrmOutputManager

[cmeissl]

highly wip of the ideas described here: pop-os/cosmic-comp#969 (comment)

it implements most of the stuff, except:

• the capabilities stuff
• the actual format selection
• some things are stubs and need to be implemented properly

not extensively tested, but anvil seems to still be able to launch.
returning an error from DrmCompositor::new also successfully triggers the format selection
logic

TODO

A lot...

• Verify that this does not mess up the render logic with multiple commits triggering vblanks
  If this turns out to be problematic we can disable the event for commit_frame, but we need to make sure
  to kick off rendering in this case somehow..

[Drakulix]

TODO

A lot...

* [ ]  Verify that this does not mess up the render logic with multiple commits triggering vblanks
  If this turns out to be problematic we can disable the event for `commit_frame`, but we need to make sure
  to kick off rendering in this case somehow..

Addressed in 3fe19dd. While it works okay-ish for anvil, it definitely would mess up state tracking in niri and cosmic-comp and while potentially work-arounds exists, at least in the latter with multi-threading this requires ugly synchronization, while the point of this api is to hide that.

[Drakulix]

All the code paths seem to now work pretty nicely in pop-os/cosmic-comp#1020 (though I don't have any devices that actually have to use these paths, I just forced them manually). This just needs a new pixman-rs release to pass tests.

[cmeissl]

So, in general I believe the changes, except the one limiting the primary format, are pretty harmless. The DrmOutputManager is a really nice optional addition which will definitely need some more work in the future. But I feel like this provides a solid base to do so. I am really happy how this turned out, thanks!

I really like the approach with DrmOutputRenderElements, nice work :)

[cmeissl]

This needs to be configurable from the outside, I am just not sure how and in what detail. Having this restriction will kill performance on lower end hardware. In my case I would only enable this on general purpose x68 hardware and only when more than a single output is enabled.

[Drakulix]

Hmm, maybe just have this be another bit for FrameMode? We might also have a need for a new flag here. So maybe FrameMode should become FrameFlags and have something like ALLOW_PRIMARY_SCANOUT instead of PRIMARY_SCANOUT.

[Drakulix]

I have pushed a commit doing this: 54379b2

[cmeissl]

pixman-rs 0.2.1 is out