[DebugInfo][RemoveDIs] Adjust AMDGPU passes to work with DPValues (#78736)

This patch tweaks two AMDGPU passes to use iterators rather than
instruction pointers for expressing an insertion point. This is needed
to accurately support DPValues, the non-instruction storage object for
debug-info.

Two tests were sensitive to this change (variable assignments were being
put in the wrong place), and I've added extra run-lines with the "try
new debug-info..." flag. These get tested on our public buildbot to
ensure they continue to work accurately.

Author: jmorse

llvm/lib/Target/AMDGPU/AMDGPULowerKernelArguments.cpp

@@ -106,7 +106,7 @@ static bool lowerKernelArguments(Function &F, const TargetMachine &TM) {
   LLVMContext &Ctx = F.getParent()->getContext();
   const DataLayout &DL = F.getParent()->getDataLayout();
   BasicBlock &EntryBlock = *F.begin();
-  IRBuilder<> Builder(&*getInsertPt(EntryBlock));
+  IRBuilder<> Builder(&EntryBlock, getInsertPt(EntryBlock));
 
   const Align KernArgBaseAlign(16); // FIXME: Increase if necessary
   const uint64_t BaseOffset = ST.getExplicitKernelArgOffset();

llvm/lib/Target/AMDGPU/SIAnnotateControlFlow.cpp

@@ -329,15 +329,16 @@ bool SIAnnotateControlFlow::closeControlFlow(BasicBlock *BB) {
   }
 
   Value *Exec = popSaved();
-  Instruction *FirstInsertionPt = &*BB->getFirstInsertionPt();
+  BasicBlock::iterator FirstInsertionPt = BB->getFirstInsertionPt();
   if (!isa<UndefValue>(Exec) && !isa<UnreachableInst>(FirstInsertionPt)) {
     Instruction *ExecDef = cast<Instruction>(Exec);
     BasicBlock *DefBB = ExecDef->getParent();
     if (!DT->dominates(DefBB, BB)) {
       // Split edge to make Def dominate Use
-      FirstInsertionPt = &*SplitEdge(DefBB, BB, DT, LI)->getFirstInsertionPt();
+      FirstInsertionPt = SplitEdge(DefBB, BB, DT, LI)->getFirstInsertionPt();
     }
-    IRBuilder<>(FirstInsertionPt).CreateCall(EndCf, {Exec});
+    IRBuilder<>(FirstInsertionPt->getParent(), FirstInsertionPt)
+        .CreateCall(EndCf, {Exec});
   }
 
   return true;

llvm/test/CodeGen/AMDGPU/llvm.dbg.value.ll

@@ -1,6 +1,9 @@
 ; RUN: llc -O0 -mtriple=amdgcn-unknown-amdhsa -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,NOOPT %s
 ; RUN: llc -mtriple=amdgcn-unknown-amdhsa -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,OPT %s
 
+; RUN: llc -O0 -mtriple=amdgcn-unknown-amdhsa -verify-machineinstrs < %s --try-experimental-debuginfo-iterators | FileCheck -check-prefixes=GCN,NOOPT %s
+; RUN: llc -mtriple=amdgcn-unknown-amdhsa -verify-machineinstrs < %s --try-experimental-debuginfo-iterators | FileCheck -check-prefixes=GCN,OPT %s
+
 ; GCN-LABEL: {{^}}test_debug_value:
 ; NOOPT: .loc	1 1 42 prologue_end     ; /tmp/test_debug_value.cl:1:42
 ; NOOPT-NEXT: s_load_dwordx2 s[4:5], s[6:7], 0x0

llvm/test/CodeGen/AMDGPU/si-annotate-dbg-info.ll

@@ -1,5 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
 ; RUN: opt -mtriple=amdgcn-- -S -structurizecfg -si-annotate-control-flow %s | FileCheck -check-prefix=OPT %s
+; RUN: opt -mtriple=amdgcn-- -S -structurizecfg -si-annotate-control-flow %s --try-experimental-debuginfo-iterators | FileCheck -check-prefix=OPT %s
 
 define amdgpu_ps i32 @if_else(i32 %0) !dbg !5 {
 ; OPT-LABEL: define amdgpu_ps i32 @if_else(