Feature/backport_ea1_ops #105
Draft
Conversation
This file contains 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
Closed
…ENABLE_PYTORCH_EXTENSIONS Disable test for TORCH_MLIR_ENABLE_PYTORCH_EXTENSIONS
Added `backend-legal-ops` argument in `fx.import_and_export` to stop decomposition of certain torch ops. This PR is based on this [issue](llvm#3953)
Bumps [externals/llvm-project](https://github.com/Xilinx/llvm-project) from `f670e5d` to `c86e97b`. - [Commits](Xilinx/llvm-project@f670e5d...c86e97b) --- updated-dependencies: - dependency-name: externals/llvm-project dependency-type: direct:production ... Signed-off-by: dependabot[bot] <[email protected]>
…lvm-project-c86e97b Bump externals/llvm-project from `f670e5d` to `c86e97b`
Drops Python 3.8, which is EOL, see https://devguide.python.org/versions/, and adds Python 3.12, which is the default e.g. on Ubuntu 24.04 LTS.
[AutoBump] Merge with fixes of edd1bbe (Oct 14, needs LLVM bump to Oct11) (82)
…lvm#3946) The behavior of float -> integer cast in PyTorch (though I haven't found the actual code implementing the cast) appears to be (based on the results produced in PyTorch): 1. round the float nearest to zero (similar to `arith.fptosi/ui`) 2. then perform the conversion Currently we only emit `tosa.cast` for this operation but as per the spec https://www.mlplatform.org/tosa/tosa_spec.html#_cast the rounding performed for float -> integer is round to nearest integer (not zero). Hence, the current TOSA lowering for `torch.ops.aten.to` produces incorrect answer.
When center_point_box=1, the supplied boxes come with a format of [x_center, y_center, width, height], this patch converts the format into [x1, y1, x2, y2] format before they are consumed. The e2e test is added in nod-ai/SHARK-TestSuite#436
[AutoBump] Merge with 895f490 (Oct 15) (83)
Bumps [externals/llvm-project](https://github.com/Xilinx/llvm-project) from `ca3473c` to `951cb07`. - [Commits](Xilinx/llvm-project@ca3473c...951cb07) --- updated-dependencies: - dependency-name: externals/llvm-project dependency-type: direct:production ... Signed-off-by: dependabot[bot] <[email protected]>
…lvm-project-951cb07 Bump externals/llvm-project from `ca3473c` to `951cb07`
[AutoBump] Merge with dc7a1ff (Oct 16) (85)
[AutoBump] Merge with 09cdbe4 (Oct 18) (87)
[AutoBump] Merge with fixes of 45bb17e (Oct 15) (84)
[AutoBump] Merge with fixes of f5d15ab (Oct 20) (88)
This commit sets the PyTorch and TorchVision version to nightly release 2025-01-20. This commit also adds the aten::_assert_tensor_metadata op by adding a folder for the op. Signed-off-by: Vivek Khandelwal <[email protected]>
Bumps [externals/llvm-project](https://github.com/Xilinx/llvm-project) from `5346a56` to `479c8d6`. - [Commits](Xilinx/llvm-project@5346a56...479c8d6) --- updated-dependencies: - dependency-name: externals/llvm-project dependency-type: direct:production ... Signed-off-by: dependabot[bot] <[email protected]>
[AutoBump] Merge with fixes of 9a167e2 (Jan 10) (151)
[AutoBump] Merge with 62eb38b (Jan 14) (152)
[AutoBump] Merge with fixes of 040aec9 (Jan 14) (153)
[AutoBump] Merge with 09af3b6 (Jan 14) (154)
…lvm-project-479c8d6 Bump externals/llvm-project from `5346a56` to `479c8d6`
[AutoBump] Merge with fixes of 33337fc (Jan 17) (155)
[AutoBump] Merge with db82bb9 (Jan 17) (156)
[AutoBump] Merge with fixes of 8fa3bd9 (Jan 17) (157)
[AutoBump] Merge with af8514c (Jan 27) (160)
[AutoBump] Merge with fixes of 13ee7c2 (Dec 19) (143)
[AutoBump] Merge with 9dd94fb (Jan 22) (158)
[AutoBump] Merge with fixes of 481da8d (Jan 22) (159)
Bumps [externals/llvm-project](https://github.com/Xilinx/llvm-project) from `479c8d6` to `e69f389`. - [Commits](Xilinx/llvm-project@479c8d6...e69f389) --- updated-dependencies: - dependency-name: externals/llvm-project dependency-type: direct:production ... Signed-off-by: dependabot[bot] <[email protected]>
…lvm-project-e69f389 Bump externals/llvm-project from `479c8d6` to `e69f389`
Bumps [externals/llvm-project](https://github.com/Xilinx/llvm-project) from `e69f389` to `91ca2e1`. - [Commits](Xilinx/llvm-project@e69f389...91ca2e1) --- updated-dependencies: - dependency-name: externals/llvm-project dependency-type: direct:production ... Signed-off-by: dependabot[bot] <[email protected]>
…lvm-project-91ca2e1 Bump externals/llvm-project from `e69f389` to `91ca2e1`
Bumps [externals/llvm-project](https://github.com/Xilinx/llvm-project) from `91ca2e1` to `2b2e860`. - [Commits](Xilinx/llvm-project@91ca2e1...2b2e860) --- updated-dependencies: - dependency-name: externals/llvm-project dependency-type: direct:production ... Signed-off-by: dependabot[bot] <[email protected]>
…lvm-project-2b2e860 Bump externals/llvm-project from `91ca2e1` to `2b2e860`
Those changes do not affect tests; upstream seems to have other means to make things pass.
Reduce diff with respect to upstream
Reduce diff to upstream
Bumps [externals/llvm-project](https://github.com/Xilinx/llvm-project) from `2b2e860` to `1656bbb`. - [Commits](Xilinx/llvm-project@2b2e860...1656bbb) --- updated-dependencies: - dependency-name: externals/llvm-project dependency-type: direct:production ... Signed-off-by: dependabot[bot] <[email protected]>
…lvm-project-1656bbb Bump externals/llvm-project from `2b2e860` to `1656bbb`
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.
This is not to be merged; just an easy way to see the accumulated changes.