[RFC] Support Codegen Using Texture Memory

Current TVM is not supporting mixture texture and buffer. This pattern is essential to obtain state-of-art performance on iOS device, and potentially better OpenGL backend for legacy devices.

A potential solution is providing extra device hints, eg:

a = tvm.placeholder((32,), name="a", storage_dev="2d_texture")
b = tvm.placeholder((32,), name="b", storage_dev="buffer")

In lowering process, some extra work is required I think:

1. We need a pass to infer output storage dev for each node
2. We need special data access lowing rule for texture, especially different layout of texture (1D/2D texture)
3. We need modify codegen
4. We need to modify runtime, to support NDArray laying on texture memory.

Once we support 2D texture, I think we are able to achieve state-of-art performance on iOS, and OpenGL.

[masahi]

Thanks @tmoreau89, I’ve opened a fresh tracking https://github.com/apache/tvm/issues/9891

[tmoreau89]

Good idea, thank you @masahi !