Update dependencies#191
Open
ypapadop-amd wants to merge 5 commits into
Open
Conversation
There was a problem hiding this comment.
Pull request overview
Updates ggml-hsa’s Python dependency set and aligns IRON kernel generation code with upstream IRON API changes.
Changes:
- Bump
mlir_aiedependency from 1.3.1 to 1.3.3 and adjust the associated extra index URL. - Remove explicit
SequentialPlacerusage and rely on the defaultProgram.resolve_program()behavior across IRON kernels. - Move
ruffinto a newrequirements-dev.txtand adjust kernel backend preference ordering forGGML_OP_ADD.
Reviewed changes
Copilot reviewed 12 out of 12 changed files in this pull request and generated 3 comments.
Show a summary per file
| File | Description |
|---|---|
| src/ggml-hsa/requirements-triton.txt | Removes ruff from Triton environment requirements. |
| src/ggml-hsa/requirements-iron.txt | Updates MLIR-AIE index + pins mlir_aie==1.3.3; removes ruff from base env. |
| src/ggml-hsa/requirements-dev.txt | Introduces a dev-only requirements file containing ruff. |
| src/ggml-hsa/kernels/iron/unary_ops.py | Drops SequentialPlacer import and uses default resolve_program(). |
| src/ggml-hsa/kernels/iron/softmax.py | Drops SequentialPlacer import and uses default resolve_program() in all program constructors. |
| src/ggml-hsa/kernels/iron/scale.py | Drops SequentialPlacer import and uses default resolve_program(). |
| src/ggml-hsa/kernels/iron/cross_entropy_loss.py | Drops SequentialPlacer import and uses default resolve_program(). |
| src/ggml-hsa/kernels/iron/count_equal.py | Drops SequentialPlacer import and uses default resolve_program(). |
| src/ggml-hsa/kernels/iron/clamp.py | Drops SequentialPlacer import and uses default resolve_program(). |
| src/ggml-hsa/kernels/iron/binary_ops.py | Drops SequentialPlacer import and uses default resolve_program() in both paths. |
| src/ggml-hsa/kernels/iron/argmax.py | Drops SequentialPlacer import and uses default resolve_program(). |
| src/ggml-hsa/kernels/binary_ops.py | Changes the backend attempt order for ggml_op_add() (TRITON vs IRON ordering). |
Comment on lines
157
to
162
| return [ | ||
| _make_triton_add_kernel_spec(arch, input_tensors, output_tensor), | ||
| _make_iron_binary_kernel_spec( | ||
| arch, input_tensors, output_tensor, "GGML_OP_ADD" | ||
| ), | ||
| _make_triton_add_kernel_spec(arch, input_tensors, output_tensor), | ||
| ] |
Comment on lines
+5
to
10
| --extra-index-url https://github.com/Xilinx/mlir-aie/releases/expanded_assets/v1.3.3 | ||
| --extra-index-url https://github.com/Xilinx/llvm-aie/releases/expanded_assets/nightly | ||
| --extra-index-url https://pypi.org/simple | ||
|
|
||
| mlir_aie==1.3.1 | ||
| mlir_aie==1.3.3 | ||
| llvm-aie |
Comment on lines
+1
to
+3
| # Copyright (c) 2026 Advanced Micro Devices, Inc. All Rights Reserved. | ||
|
|
||
| ruff |
Comment on lines
+5
to
+9
| --extra-index-url https://github.com/Xilinx/mlir-aie/releases/expanded_assets/v1.3.3 | ||
| --extra-index-url https://github.com/Xilinx/llvm-aie/releases/expanded_assets/nightly | ||
| --extra-index-url https://pypi.org/simple | ||
|
|
||
| mlir_aie==1.3.1 | ||
| mlir_aie==1.3.3 |
Comment on lines
+1
to
+3
| # Copyright (c) 2026 Advanced Micro Devices, Inc. All Rights Reserved. | ||
|
|
||
| ruff |
Comment on lines
+38
to
+42
| # If both iron and triton are selected, only install triton requirements | ||
| INSTALL_LIST=("${BACKEND_LIST[@]}") | ||
| if [[ " ${BACKEND_LIST[*]} " == *" iron "* ]] && [[ " ${BACKEND_LIST[*]} " == *" triton "* ]]; then | ||
| INSTALL_LIST=("triton") | ||
| fi |
…riton when both backends are selected
… Triton compilation process
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.
No description provided.