[PATCH] D148571: [coro][nfc] Use Load/Store instruction APIs

[Felipe de Azevedo Piovezan via Phabricator via llvm-commits]

fdeazeve created this revision.
fdeazeve added a reviewer: aprantl.
Herald added subscribers: ChuanqiXu, hiraditya.
Herald added a project: All.
fdeazeve requested review of this revision.
Herald added a project: LLVM.
Herald added a subscriber: llvm-commits.

Instead of hard-coding operand indices, it is cleaner to use the APIs
provided by the Load/Store Inst classes.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D148571

Files:
  llvm/lib/Transforms/Coroutines/CoroFrame.cpp


Index: llvm/lib/Transforms/Coroutines/CoroFrame.cpp
===================================================================
--- llvm/lib/Transforms/Coroutines/CoroFrame.cpp
+++ llvm/lib/Transforms/Coroutines/CoroFrame.cpp
@@ -2832,7 +2832,7 @@
 
   while (auto *Inst = dyn_cast_or_null<Instruction>(Storage)) {
     if (auto *LdInst = dyn_cast<LoadInst>(Inst)) {
-      Storage = LdInst->getOperand(0);
+      Storage = LdInst->getPointerOperand();
       // FIXME: This is a heuristic that works around the fact that
       // LLVM IR debug intrinsics cannot yet distinguish between
       // memory and value locations: Because a dbg.declare(alloca) is
@@ -2842,7 +2842,7 @@
       if (!SkipOutermostLoad)
         Expr = DIExpression::prepend(Expr, DIExpression::DerefBefore);
     } else if (auto *StInst = dyn_cast<StoreInst>(Inst)) {
-      Storage = StInst->getOperand(0);
+      Storage = StInst->getValueOperand();
     } else {
       SmallVector<uint64_t, 16> Ops;
       SmallVector<Value *, 0> AdditionalValues;


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D148571.514402.patch
Type: text/x-patch
Size: 1021 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20230417/0c35c94a/attachment.bin>