[PATCH] D154019: [IR] Ignore globals with the `llvm.` prefix when calculating module hash

[Arthur Eubanks via Phabricator via llvm-commits]

aeubanks added inline comments.


================
Comment at: llvm/unittests/IR/PassManagerTest.cpp:1090
+
+TEST_F(PassManagerTest, ModulePassNoAnalysisInvalidation) {
+  LLVMContext Context;
----------------
paulkirth wrote:
> aeubanks wrote:
> > I've been meaning to write some unittests that specifically just check (in)equality of `llvm::StructuralHash` of two modules so we can avoid doing that. Going through a pass manager shouldn't be necessary. I can do that, or if you have time you can start that :)
> I'm going on vacation starting tomorrow,so I'm not sure I'll have time to get to it until after that.
> 
> But, just in case I do scrounge the time, are there other important cases you know of of the top of your head?
it's basically just all the cases in StructuralHashImpl, there aren't that many right now. it would be good to get some tests out so it's easy to test future changes


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D154019/new/

https://reviews.llvm.org/D154019