Repository navigation
Conversation
This was referenced Nov 15, 2026
Contributor
Author
|
@tqchen, could you review this small follow-up to #20976? The unchanged pinned TIRx commit now has the final v0.1.0 tag, so source wheel builds resolve to 0.1.0 and the parent rejects them under its rc2/rc3-only bounds. Both macOS and Windows hit this before tests across all five refreshed PRs linked above. The fix accepts the final release in CMake and both Python dependency bounds, while keeping the source pin and the upper limit at 0.1.0. Installed-package configuration and generated metadata validation pass locally; native CI has been triggered. Thank you. |
tlopex
pushed a commit
that referenced
this pull request
Nov 15, 2026
With rank-one weights, `take(..., axis=0)` selects elements on the MatMul reduction axis. Moving it after MatMul raises on vector activations and changes results on matrix activations, even when the output shape stays the same. Require weight rank >= 2 in the simple column-selection rewrite. Leave vector weights unchanged. Add two regressions and a positive control showing that vector activations with matrix weights still rewrite. Fixes #20595. Validation against main `30fd3017fd`, Python 3.11, LLVM 21.1.8 and source-built pinned TIRx/FFI dependencies: - `test_transform_reorder_take_after_matmul.py`, `test_transform_expand_matmul_of_sum.py`, and `test_op_linear_algebra.py`: 26 passed. - CPU C-backend VM execution against NumPy: 20/20 transformed cases match values, shapes and dtypes. Covers vector/matrix/batched activations, symbolic batch dimensions, and fast/clip/wrap take modes. All 20 original modules also match the reference. - File-scoped pre-commit checks passed. Prepared with OpenAI Codex assistance. The build and validation above were executed locally. Includes the shared packaging fix in #20980: the pinned TIRx source now carries the final `0.1.0` tag, which the old CMake/Python bounds rejected on macOS and Windows. The pass/operator implementation is unchanged by this CI follow-up. Fresh CI results are pending.
Member
|
Closed as already included in #20815 |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The pinned TIRx commit
f43b3030d33196d3a5f21e56b37549a6d3a5c70bnow carries thev0.1.0release tag. After CI fetches tags, its source-built wheel isapache-tvm-tirx==0.1.0. TVM still accepts only0.1.0rc2/0.1.0rc3, so the parent wheel fails to configure on both macOS and Windows before running tests:Examples: macOS job, Windows job. The same error occurs in all five refreshed PRs #20815, #20816, #20817, #20818 and #20819.
Accept the final
0.1.0release in the installed-package CMake check and align the build/runtime Python dependency bounds. Keep support for rc2/rc3 and cap the dependency at0.1.0; later patch/minor versions remain excluded. The pinned TIRx source commit is unchanged.Validation on Linux aarch64:
0.1.0: unchanged main fails with the CI error; the patched tree configures successfully.Requires-Distaccepts0.1.0and rejects0.1.1.Native macOS/Windows wheel builds are left to CI; local validation covers installed-package configuration and generated metadata.
Prepared with OpenAI Codex assistance.