Initial work for WebGPU and refactor GPU namespace #11
Conversation
|
|
||
| "include/tempeh/gpu/vk/device.hpp" | ||
| "include/tempeh/gpu/vk/vk.hpp" | ||
| "include/tempeh/gpu/vk/volk.h") |
There was a problem hiding this comment.
I revert this changes and will never merge this to the master. If you are using the master branch as your base for the implementation, it's incomplete. Please use vulkan-impl branch as the base of WebGPU backend implementation, it contains complete (almost) interfaces.
| #include "vk.hpp" | ||
|
|
||
| namespace Tempeh::GPU | ||
| namespace Tempeh::GPU::Vk |
There was a problem hiding this comment.
Do not add too many namespaces.
Seriously, you can always put the name of the backend at the end of the derived class name.
Deep namespace is unnecessary in this case.
These classes are not going to be visible to the end-user anyway, so adding the namespace is completely useless.
There was a problem hiding this comment.
Namespace good. If its too long, you can just use alias
There was a problem hiding this comment.
Not good, alias can add redundant lines. It's just not worth it.
Why add deep namespaces when can just DeviceVK. Tbh, it's not worth it in this case.
| using Box = std::unique_ptr<T>; | ||
|
|
||
| template<typename T> | ||
| FUNCTION_ALIAS(make_rc, std::make_shared<T>); |
There was a problem hiding this comment.
make_shared & make_unique requires you to provide the constructor parameters, this is probably wont work for classes that don't have a default constructor.
No description provided.