Skip to content

present/wnmd: Only queue vblank event if target MSC is in the future

If we always queue for current MSC + 1, it means apps explicitly waiting for the swap completion can never achieve frame-rates higher than the refresh rate, even if they're trying to swap buffers as quickly as possible, e.g. via using glXSwapBuffersMscOML(..., 0, 0, 0) or setting vblank_mode=0.

Closes: #1103 Fixes: 1cccb486 "present/wnmd: Execute copies at target_msc-1 already"

Merge request reports