[DAG] SDPatternMatch - m_Reassociatable mismatches

[RKSimon]

Noticed while working on #169644

If we attempt to match a m_ReassociatableAdd pattern containing both the m_Value and m_Deferred for a given SDValue, the matcher gets corrupted and fails to match the correct SDValue pairs

foldAddToAvg:

  if ((!LegalOperations || hasOperation(ISD::AVGFLOORU, VT)) &&
      (sd_match(N,
                m_Add(m_And(m_Value(A), m_Value(B)),
                      m_Srl(m_Xor(m_Deferred(A), m_Deferred(B)), m_One()))) ||
       sd_match(N, m_ReassociatableAdd(
                       m_Srl(m_Value(A), m_One()), m_Srl(m_Value(B), m_One()),
                       m_ReassociatableAnd(m_Deferred(A), m_Deferred(B),
                                           m_One()))))) {
    return DAG.getNode(ISD::AVGFLOORU, DL, VT, A, B);
  }

in this case we end up with the same SDValue being matched for A and B.

Hopefully this is just an evaluation ordering issue, but there might be something else going on.

Good first issue for someone who likes template programming!

[llvmbot]

Hi!

This issue may be a good introductory issue for people new to working on LLVM. If you would like to work on this issue, your first steps are:

1. Check that no other contributor is working on this issue. If someone is assigned to the issue or claimed to be working on it, ping the person. After one week without a response, the assignee may be changed.
2. Leave a comment indicating that you are working on the issue, or just create a pull request after following the steps below. Mention this issue in the description of the pull request.
3. Fix the issue locally.
4. Run the test suite locally. Remember that the subdirectories under test/ create fine-grained testing targets, so you can e.g. use make check-clang-ast to only run Clang's AST tests.
5. Create a Git commit.
6. Run git clang-format HEAD~1 to format your changes.
7. Open a pull request to the upstream repository on GitHub. Detailed instructions can be found in GitHub's documentation. Mention this issue in the description of the pull request.

If you have any further questions about this issue, don't hesitate to ask via a comment in the thread below.

[llvmbot]

@llvm/issue-subscribers-good-first-issue

Author: Simon Pilgrim (RKSimon)

Noticed while working on #169644

If we attempt to match a m_ReassociatableAdd pattern containing both the m_Value and m_Deferred for a given SDValue, the matcher gets corrupted and fails to match the correct SDValue pairs

foldAddToAvg:

  if ((!LegalOperations || hasOperation(ISD::AVGFLOORU, VT)) &&
      (sd_match(N,
                m_Add(m_And(m_Value(A), m_Value(B)),
                      m_Srl(m_Xor(m_Deferred(A), m_Deferred(B)), m_One()))) ||
       sd_match(N, m_ReassociatableAdd(
                       m_Srl(m_Value(A), m_One()), m_Srl(m_Value(B), m_One()),
                       m_ReassociatableAnd(m_Deferred(A), m_Deferred(B),
                                           m_One()))))) {
    return DAG.getNode(ISD::AVGFLOORU, DL, VT, A, B);
  }

in this case we end up with the same SDValue being matched for A and B.

Hopefully this is just an evaluation ordering issue, but there might be something else going on.

Good first issue for someone who likes template programming!

[bermondd]

Hi! I wanna work on this issue. Can I be assigned to it?

[RKSimon]

@bermondd sorry I assigned this to me by mistake - it's all yours