I notice some more suspicious swapchain RFC keyword usage:

The possible return values for vkGetSwapchainStatusKHR should be interpreted as follows:

The "should" feels odd here and colloquial. Should be refrazed in the form of what the implementation must return.

If a counter is not available because the swapchain is out of date, the implementation may return VK_ERROR_OUT_OF_DATE_KHR.

"may" feels suspicious here. Posibly supposed to be "must".

As usual, vkQueuePresentKHR may fail if oldSwapchain has entered a state that causes VK_ERROR_OUT_OF_DATE_KHR to be returned.

The "may" is weird, because tautologically if VK_ERROR_OUT_OF_DATE_KHR is returned, then it did fail.