[llvm] [AMDGPU] Optimizing Dynamic Alloca S-DAG I-Sel (PR #124292)

[via llvm-commits]

easyonaadit wrote:

> > dyn_size = max_size << wave_size
> 
> Here dyn_size = max_size << 5/6 right ? not `wave_size` ?

Right, I've updated it. 

https://github.com/llvm/llvm-project/pull/124292