For reference: https://vulkan.org/user/pages/09.events/vulkanised-2025/T47-...

coopmat2 features will eventually be rolled elsewhere. coopmat also started as an NVIDIA extension.

The client use cases that coopmat was intended for are customer machines, not multi-GPU, which is broadly seen as a datacenter feature instead. That said coopmat orthogonal to this.

So when I said "a couple of nvidia engineers", I indeed meant Jeff.

VK_KHR_cooperative_matrix - embrace?

VK_NV_cooperative_matrix2 - extend?

I am pretty sure VkImportSemaphoreFdInfoKHR, mentioned in https://github.com/ggml-org/llama.cpp/issues/22648, works across multiple AMD devices, but somehow doesn't work across multiple nvidia devices.

> I am pretty sure VkImportSemaphoreFdInfoKHR, mentioned in https://github.com/ggml-org/llama.cpp/issues/22648, works across multiple AMD devices, but somehow doesn't work across multiple nvidia devices.

p2p is disabled on nvidia customer cards, vulkan device groups are shipped for the RTX 6000s

> Added support for creating Vulkan logical devices from multiple physical devices on select cards via VK_KHR_device_group_creation. This feature can be enabled by setting the environment variable __VK_ENABLE_DEVICE_GROUPS=1.

Back to the topic about cuda moat, in the slide with title "Problems with Coopmat1", the current frontier open models have absolutely no issue with:

* manual pipelining

* shared memory staging

* tiling

* bounds checking

A big problem there is ensuring performance portability between different GPUs