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

[Pravin Jagtap via llvm-commits]

pravinjagtap wrote:

> dyn_size = max_size << wave_size

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

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