Jeffrey Knockel

It looks like it's not just applications that sometimes assume that the returned X11 display modes are equivalent to logical coordinates. I noticed that SDL itself in at least one...

> Everything makes sense including the reasoning for the fix, so my only question is less about the change and more about how this interacts with X itself: I always...

The issue may be related to the one described here: https://bugs.launchpad.net/ubuntu/+source/glibc/+bug/1944468/comments/9 If so, the issue appears fixed in chromium as of June 2021: https://chromium.googlesource.com/chromium/src/+/218438259dd795456f0a48f67cbe5b4e520db88b From what I can tell, Wire...

Hi @matt335672 , I am experiencing a similar issue as @mustafa58 . Adding `Option "AutoAddGPU" "off"` as you suggested didn't work for me. However, I found that adding `GPUDevice ""`...

> Now that SDL 3.0's high DPI support is ready, we can revisit this. Since the scaled modes affect all X11 coordinates, we can accept this patch as-is (once the...

I've managed to convince myself that `eglWaitClient()` or some analog is not required before resizing. For one thing, I can't reproduce any race without that synchronization. Furthermore, there are many...

I just realized that this approach breaks transparent backgrounds. But perhaps if I set the parent window to be fully transparent...

> I just realized that this approach breaks transparent backgrounds. But perhaps if I set the parent window to be fully transparent... Yes this appears to work.

Thanks, adding `Option "DisableVidModeExtension" "on"` to the ServerFlags section prevents the crash on my end as well.

I have built some wheels using mingw/msys against both python 3.10 and 3.11 here: https://www.cs.unm.edu/~jeffk/jq/ These have had very little testing, so please use at your own risk. To get...