[llvm] [AMDGPU]Optimize SGPR spills (PR #93668)

[Matt Arsenault via llvm-commits]

arsenm wrote:

> > [kernel-hip-amdgcn-amd-amdhsa-gfx908.zip](https://github.com/user-attachments/files/15905813/kernel-hip-amdgcn-amd-amdhsa-gfx908.zip)
> 
> > 
> 
> > To accurately reproduce the assembly file, use the following clang command:
> 
> > 
> 
> > ```
> 
> > clang -cc1 -mcode-object-version=4 -mllvm --amdhsa-code-object-version=4 -triple amdgcn-amd-amdhsa -aux-triple x86_64-unknown-linux-gnu -S -disable-free -mrelocation-model pic -pic-level 2 -mframe-pointer=none -aux-target-cpu x86-64 -mllvm -amdgpu-internalize-symbols -target-cpu gfx908 -O3 -vectorize-loops -vectorize-slp -faddrsig -o tmp.s -x ir kernel-hip-amdgcn-amd-amdhsa-gfx908.bc
> 
> > ```
> 
> 
> 
> Getting this while trying to reproduce.
> 
> 
> 
> ![dumpError](https://github.com/llvm/llvm-project/assets/35700483/f09aa238-13a4-4fd9-b570-0d0606096138)
> 
> 

This is due to mixing upstream and amd-staging with the different debug info 

https://github.com/llvm/llvm-project/pull/93668