[llvm] [NVPTX] Implement computeKnownBitsForTargetNode for LoadV2/4 (PR #154165)

[Alex MacLean via llvm-commits]

https://github.com/AlexMaclean commented:

Nice this looks simpler. I'd prefer that we get even a little more generic still though. I think currently the only time we'll ever generate extending vector loads is for i8 but that might change in the future and I think it would not be too complex to implement a generic version of this function which does something like the following: get the size in bits of the memVT and divide that by the number of elements loaded then get the difference between that value and the actual size of the produced VT and clear that many high bits. 

https://github.com/llvm/llvm-project/pull/154165