[PATCH] D64261: [CodeGen] Enhance `MachineInstrSpan` to allow the end of MBB to be used.

Tue Jul 9 07:27:29 PDT 2019

dstenb added a comment.

The tests introduced in this commit currently fail when running UBSan, due to invoking `getRegInfo()` with `RegInfo` being null:

  [ RUN      ] MachineInstrSpan.DistanceBegin
  /b/sanitizer-x86_64-linux-fast/build/llvm/include/llvm/CodeGen/MachineFunction.h:474:46: runtime error: reference binding to null pointer of type 'llvm::MachineRegisterInfo'
      #0 0x1bed93c in getRegInfo /b/sanitizer-x86_64-linux-fast/build/llvm/include/llvm/CodeGen/MachineFunction.h:474:39
      #1 0x1bed93c in llvm::ilist_traits<llvm::MachineInstr>::addNodeToList(llvm::MachineInstr*) /b/sanitizer-x86_64-linux-fast/build/llvm/lib/CodeGen/MachineBasicBlock.cpp:111
      #2 0x48939c in llvm::iplist_impl<llvm::simple_ilist<llvm::MachineInstr, llvm::ilist_sentinel_tracking<true> >, llvm::ilist_traits<llvm::MachineInstr> >::insert(llvm::ilist_iterator<llvm::ilist_detail::node_options<llvm::MachineInstr, true, true, void>, false, false>, llvm::MachineInstr*) /b/sanitizer-x86_64-linux-fast/build/llvm/include/llvm/ADT/ilist.h:227:11
      #3 0x488e2a in llvm::MachineBasicBlock::insert(llvm::MachineInstrBundleIterator<llvm::MachineInstr, false>, llvm::MachineInstr*) /b/sanitizer-x86_64-linux-fast/build/llvm/include/llvm/CodeGen/MachineBasicBlock.h:627:18
      #4 0x488a93 in (anonymous namespace)::MachineInstrSpan_DistanceBegin_Test::TestBody() /b/sanitizer-x86_64-linux-fast/build/llvm/unittests/CodeGen/MachineInstrTest.cpp:289:8
      #5 0x2506006 in testing::Test::Run() /b/sanitizer-x86_64-linux-fast/build/llvm/utils/unittest/googletest/src/gtest.cc:2474:5
      #6 0x2506cf3 in testing::TestInfo::Run() /b/sanitizer-x86_64-linux-fast/build/llvm/utils/unittest/googletest/src/gtest.cc:2656:11
      #7 0x25076d2 in testing::TestCase::Run() /b/sanitizer-x86_64-linux-fast/build/llvm/utils/unittest/googletest/src/gtest.cc:2774:28
      #8 0x250ee02 in testing::internal::UnitTestImpl::RunAllTests() /b/sanitizer-x86_64-linux-fast/build/llvm/utils/unittest/googletest/src/gtest.cc:4649:43
      #9 0x250e846 in testing::UnitTest::Run() /b/sanitizer-x86_64-linux-fast/build/llvm/utils/unittest/googletest/src/gtest.cc:4257:10
      #10 0x492765 in RUN_ALL_TESTS /b/sanitizer-x86_64-linux-fast/build/llvm/utils/unittest/googletest/include/gtest/gtest.h:2233:46
      #11 0x492765 in main /b/sanitizer-x86_64-linux-fast/build/llvm/unittests/CodeGen/TargetOptionsTest.cpp:75
      #12 0x7f0cc584c2e0 in __libc_start_main (/lib/x86_64-linux-gnu/libc.so.6+0x202e0)
      #13 0x44d429 in _start (/b/sanitizer-x86_64-linux-fast/build/llvm_build_ubsan/unittests/CodeGen/CodeGenTests+0x44d429)

Log: http://lab.llvm.org:8011/builders/sanitizer-x86_64-linux-fast/builds/33393/steps/check-llvm%20ubsan/logs/stdio

Is someone looking into that? I did not find any indication of that on llvm-commits, llvm-dev, or in bugzilla, so I'm writing on this review just to make sure.

(I have not been able to run this myself, so I have not actually verified that the tests start breaking with rL365240 <https://reviews.llvm.org/rL365240> rather than some later commit.)

Repository:
  rL LLVM

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D64261/new/

https://reviews.llvm.org/D64261