[PATCH] D155190: [amdgpu][lds] Remove recalculation of LDS frame from backend

[Jon Chesterfield via Phabricator via llvm-commits]

JonChesterfield added a comment.

This is D155125 <https://reviews.llvm.org/D155125> with the recalculation deleted, instead of set up to catch regressions under CI.



================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUMachineFunction.cpp:94
-
-static const GlobalVariable *getKernelLDSGlobalFromFunction(const Function &F) {
-  const Module *M = F.getParent();
----------------
Deleting this association gives lds lowering more data layout freedom and drops the limitation on anonymous functions


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUMachineFunction.cpp:103
 static const GlobalVariable *
 getKernelDynLDSGlobalFromFunction(const Function &F) {
   const Module *M = F.getParent();
----------------
this association will be removed in a later patch


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUMachineFunction.cpp:116
-
-void AMDGPUMachineFunction::allocateKnownAddressLDSGlobal(const Function &F) {
-  const Module *M = F.getParent();
----------------
this is the recomputation of the LDS frame being deleted


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D155190/new/

https://reviews.llvm.org/D155190