[PATCH] D155240: [amdgpu][lds] Use amdgpu-lds-size instead of llvm.donothing

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPULowerModuleLDSPass.cpp:1064
 
+          // Could replace this with a dynamic LDS alignment attribute
           markUsedByKernel(Builder, func, N);
----------------
I thought it was weird only having a size attribute. But also, how does this help if the size is always rounded by the maximum dynamic alignment?


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp:955
+
+  if (CurrentLocalMemUsage == 0) {
+    // Sort to try to estimate the worst case alignment padding
----------------
Is there now a path to introducing LDS outside of a kernel?

I'm slightly nervous about assuming amdgpu-lds-size is a source of truth in case something else introduces new LDS globals. Such as this pass, which is not updating the value 


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D155240/new/

https://reviews.llvm.org/D155240