Created attachment 1139 [details]
Patch to implement support for _NET_WM_BYPASS_COMPOSITOR

It seems like you'd want to only turn this on for fullscreen windows?

Well, the compositor should make decisions about what it can and cannot support. It's possible a compositor could redirect non-full-screen windows that are stacked to the front while it can.

(In reply to comment #3)
> Well, the compositor should make decisions about what it can and cannot
> support. It's possible a compositor could redirect non-full-screen windows
> that are stacked to the front while it can.

Yeah I have though about lifting the "must be monitor sized" restriction in mutter but only for windows that explicitly ask for being unredirected using the hint. It might make sense for windowed games but I have not tested it yet how well it works in practice but we might do that in the future.

Any update on this? Right now we're sort of using a strange hack in mutter / gnome-shell to detect 100 or so full-window damages, but that doesn't play well with some SDL2 games that resize on minimize or similar, where we get the Damage before the ConfigureNotify.

(Sorry if you get a lot of copies of this email, we're touching dozens of bug reports right now.)

Tagging a bunch of bugs as target-2.0.0, Priority 2.

This means we're in the final stretch for an official SDL 2.0.0 release! These are the bugs we really want to fix before shipping if humanly possible.

That being said, we don't promise to fix them because of this tag, we just want to make sure we don't forget to deal with them before we bless a final 2.0.0 release, and generally be organized about what we're aiming to ship.

Hopefully you'll hear more about this bug soon. If you have more information (including "this got fixed at some point, nevermind"), we would love to have you come add more information to the bug report when you have a moment.

Thanks!
--ryan.

Just making sure...

        const int _NET_WM_BYPASS_COMPOSITOR_HINT_ON = 1;

Should that be a "long"? I can never remember what XChangeProperty() wants here, and just want to make sure we get it right for both 32 and 64-bit platforms.

--ryan.

Hm, scanning through GDK, I think we do want "long" here. Can you update the patch? I can't seem to compile SDL at the moment.

Created attachment 1245 [details]
Updated patch.

(In reply to comment #8)
> Hm, scanning through GDK, I think we do want "long" here. Can you update the
> patch? I can't seem to compile SDL at the moment.

Attached. I left your name as the author of the patch, I hope you don't mind.

Sam: this seems reasonable to me. Unless you want to add a hint to allow apps to optionally disable it first, I think we should push this patch.

--ryan.

I think the only case where we want to disable compositor bypass is when the application doesn't sync to vblank, in which case you might get tearing or flickering. Of course, not syncing to vblank introduces a bunch of artifacts anyway, so it's probably not too important to change the property in that case.

(oh, and I don't mind being credited with the patch, if I have to explicitly give permission)

Applied!
http://hg.libsdl.org/SDL/rev/fc864efbc1a6