[PATCH] D100870: [MemCpyOpt] Allow variable lengths in memcpy optimizer

[Nikita Popov via Phabricator via llvm-commits]

nikic added a comment.

Looks fine, I'd suggest a few more tests:

- For the two you already have, add negative tests where the size is dynamic and doesn't match.
- Add a memcpy from uninit alloca with dynamic size.



================
Comment at: llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp:1437
                                        M->getSourceAlign().valueOrOne());
-            if (performCallSlotOptzn(M, M, M->getDest(), M->getSource(),
-                                     CopySize->getZExtValue(), Alignment, C)) {
-              LLVM_DEBUG(dbgs() << "Performed call slot optimization:\n"
-                                << "    call: " << *C << "\n"
-                                << "    memcpy: " << *M << "\n");
-              eraseInstruction(M);
-              ++NumMemCpyInstr;
-              return true;
+            if (ConstantInt *CopySize = dyn_cast<ConstantInt>(M->getLength())) {
+              if (performCallSlotOptzn(M, M, M->getDest(), M->getSource(),
----------------
I'd move this check a bit higher (accessedBetween is more expensive).


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D100870/new/

https://reviews.llvm.org/D100870