Fei Yang

Got a problem with emscripten that powf() always returns nan, including something as simple as powf(1.0f, 3.0f).

> Got a problem with emscripten that powf() always returns nan, including something as simple as powf(1.0f, 3.0f). Using "em++ -O3" seems to fix the issue. Maybe I should report...

2 points I want to mention here: * The feature is quite important for some applications. For example, rendering multiple light-sources in a forward pass, each of the light-sources has...

Yes, you are right. I recalled that the reason I had to use VK_KHR_buffer_address was because VK_KHR_ray_tracing forced me to do so in another project. So it is not the...

> This is a bug of ValidationLayers and it seems it was fixed recently. https://github.com/KhronosGroup/Vulkan-ValidationLayers/issues/1661 Yes, the validation error is gone after I updated my SDK to 1.2.141.0! It sound...

Thanks for the valuable information. Haven't been able to find P100 in the list though. Just wondering, there is not a case where VK_KHR_buffer_device_address is supported but not VK_EXT_buffer_device_address, is...

About this issue, I think I'm coming close to an ultimate solution after the last few commits. The project file now generates 2 different dll files, PyVkInline.dll and PyVkInlineEX.dll, using...

This seems to be a cffi issue. I don't see how I can fix it from ThrustRTC side. Btw, there's little benifit to multiprocess a GPU program, if any.

Yes, CURandRTC provides a device-side implementation of XORWOW algorithm. That basically copies some code from curand_kernel.h

Sorry for late reply. The only problem I see is: "class DVCupyVector" doesn't have "to_host()". Here you can do this: ```python V = cp.zeros(len(E), dtype=cp.int32) V_d = trtc.DVCupyVector(V) trtc.Exclusive_Scan(E_d, V_d)...