cracy3m

1. Thanks to reply. So I have to create Tensor object with data. 2. I just want to wrap kompute as a function called by others to do some calculations,...

> I'm not sure I understand this use-case, Tensors are currently supposed to be seen as containers of Vulkan GPU-memory data as opposed to containers of CPU-memory data, which means...

> I just had a deeper look and it does seem like there's not a trivial way in which this would be achieved as there's no concept of "uninitialised components",...

> @cracy3m I did some initial tests to validate if the change would be as straightforward as this, however there would be quite a lot of things required, namely as...

> Thank you for creating the issue @cracy3m - could you share the logs of the compile error? It may actually be due to the version of the headers, but...

@axsaucedo Is it possible that member var of class in Komputelib change private to protect (for subclass access )? I try to use Kompute src to do some gpgpu calculations...

**My test context:** platform: windows 10 (64bit) gpu: rtx 2070 vulkan sdk: 1.2.198.1 vulkan driver: I do not how to find the info, kompute version: v0.8.1 **My test code:** ```cpp...

I test `TestPushConstants.TestConstantsWrongSize` code, same error. ```cpp void test(){ { std::string shader(R"( #version 450 layout(push_constant) uniform PushConstants { float x; float y; float z; } pcs; layout (local_size_x = 1)...

@axsaucedo Thanks, I think use `vk::CommandPoolCreateFlagBits::eResetCommandBuffer` to create commandpool may avoid the validation error

Sorry, I don't have time to try the latest version recently. Maybe it would be better to close this issue.