MoltenVK is a Vulkan Portability implementation. It layers a subset of the high-performance, industry-standard Vulkan graphics and compute API over Apple's Metal graphics framework, enabling Vulkan applications to run on macOS, iOS and tvOS.
In the current implementation MoltenVK doesn't enforce a barrier when sampling a timestamp, in which case the timestamp may be sampled while some previous shaders are still running. As a result, the timings are totally unusable because they don't correctly represent the time spent on my dispatches. Is it possible to give it a YES to the withBarrier params?
In the current implementation MoltenVK doesn't enforce a barrier when sampling a timestamp, in which case the timestamp may be sampled while some previous shaders are still running. As a result, the timings are totally unusable because they don't correctly represent the time spent on my dispatches. Is it possible to give it a
YESto thewithBarrierparams?https://github.com/KhronosGroup/MoltenVK/blob/bf097edc74ec3b6dfafdcd5a38d3ce14b11952d6/MoltenVK/MoltenVK/Commands/MVKCommandBuffer.mm#L1004-L1016