dsa-t

If I move the wxGrid out of the wxNotebook, the bug doesn't seem to appear. I don't see the bug in the Bugs window of grid sample. Here's the fbp...

I think it's enough to just copy the combination of wxDialog + wxNotebook + wxGrid. The second tab (empty panel in the script) and the buttons are just to be...

I can reproduce the bug at 100% rate with the below patch. If I add `wxRESIZE_BORDER` to wxDialog style, the rate drops to 10-20%. The sizers seem to be important...

It reproduced in the wxPython script with 3.2.0 RC1 and 3.0.4. Then with the recompiled regular wxWidgets, I could't see the bug in the KiCad master branch (or I didn't...

Strangely enough, even with `3.0.4+dfsg-15build1` libs and KiCad 6.0 branch I can't reproduce the bug at this moment via VNC, nor in the modified grid sample. I've changed some vsync/tearing...

The test ``` ./graphics/bitmap.cpp:434: FAILED: CHECK( unmaskedPixelsCount == numUnmaskedPixels ) with expansion: 60 == 64 ``` fails because of the anti-aliased mask rectangle corners:  This code doesn't work because...

> Thanks for looking into this, but what would be the solution for this? I've already disabled AA for 1bpp paint devices in latest commits. But the other test also...

It works on my Linux distro, Qt prints some asserts though.

Looks like the tests passed.

> @dsa-t Let me know if you have any plan to work on making `wxQtGraphicsRenderer` available under non-Windows too ? Not in the near future.