<html> <head> <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"> </head> <body> > Indeed. The main trick is to always wait for the previous buffer swap to complete (using GLX_INTEL_swap_event or > GLX_OML_sync_control) before drawing the next frame. Then the normal _NET_WM_SYNC protocol should work. As far as I can tell I am doing that. the current loop looks something like this: <a href="https://privatebin.net/?7c57104ced0a8790#9Usgz9FTsXvTeoJmYQ5tVyMPs6sQKynb8LtXf39NnhsH" id="pasteurl" rel="nofollow noopener noreferrer">https://privatebin.net/?7c57104ced0a8790#9Usgz9FTsXvTeoJmYQ5tVyMPs6sQKynb8LtXf39NnhsH</a> (non-highlighted version at the bottom for longevity purposes) Although I might have understood wrong what you actually mean. From my PoV glXSwapBuffers should already "wait" until the frame is drawn, especially when we block afterwards using something like glClear(0) or glXWait* There's obviously a lot more code around that, but that's the basis of it. The _synced_resize is, as I mentioned, set in another thread which just polls XNextEvent. What I understand to be the problem is that during resizes we would need to render at > framerate, as resize events can be submitted at that speed, or could be submitted between vblanks asking for additional frames that the compositor needs to make things look smooth. But using VSYNC we are capped to *exactly* the framerate of the monitor that we're on, so we cannot submit frames any faster than that, especially since OpenGL must be run singlethreaded, i.e. we have no way to submit frames with VSYNC without blocking with glXSwapBuffers(). My current "fix", is to simply deactivate VSync before the buffer swap if it's a synced resize and reactivate it afterwards. This achieves quite satisfying results, although I believe it is the entirely wrong path to take. At least that way we can submit more frames than VSYNC dictates whilst keeping VSYNC on for the remainder of the program's operation. plain code version: ``` while(!_closed) { // acquire at start to avoid submitting a frame rendered for a smaller size as the response to the synced resize bool synced_resize = _synced_resize; render(); glXSwapBuffers(); // glXSwapBuffers doesn't block itself, only the next window-altering GL call does, if we want to force a block we can put a glClear(0) here if (synced_resize) { // respond to sync using counter here } } ``` </body> </html>