[Tests] Update test_adaptive_pooling_window expected IR for const-int-bound fix#20023
Conversation
…-bound fix Followup to apache#19978: the const-int-bound modular-set fix correctly prevents the simplifier from over-folding the adaptive pool window extent. The previous expected IR used the simplified closed form (v_ax2 % 3 * 4 + 16) // 12 + 1, which was only reachable because the buggy bound let CanProve prove an invalid predicate. After the fix the generated IR retains the correct T.Select form, so update the expected IR to match and remove the xfail marker that was added in apache#19978.
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request removes the @pytest.mark.xfail decorator from the test_adaptive_pooling_window test in tests/python/te/test_te_create_primfunc.py. It updates the expected TIR workload by renaming a block to adaptive_pool_sum_l1 and adjusting the loop bounds and division factor calculations to use T.Select expressions, aligning the test with the updated adaptive pool window extent behavior. There are no review comments, so I have no feedback to provide.
Important
The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.
2 participants
Followup to #19978: the const-int-bound modular-set fix correctly prevents the simplifier from over-folding the adaptive pool window extent. The previous expected IR used the simplified closed form
(v_ax2 % 3 * 4 + 16) // 12 + 1, which was only reachable because the buggy bound letCanProveprove an invalid predicate. After the fix the generated IR retains the correctT.Selectform, so update the expected IR to match and remove thexfailmarker that was added in #19978.This branch is based on current main so the
xfailremoval is explicit (addressing feedback from @tlopex on the previous attempt in #19995).