[#4585][feat] Replace unified attention before export #8303
Conversation
h-guo18
force-pushed
the
haoguo/attn_export_patch
branch
from
cec0648 to
9bb892d
Compare
📝 WalkthroughWalkthroughIntroduces a unified attention export path by registering a Torch-op-backed wrapper and patching Changes
Sequence Diagram(s)sequenceDiagram
autonumber
participant User
participant te as torch.export (te.export)
participant Patch as UnifiedAttnPatch
participant Reg as ALL_ATTENTION_FUNCTIONS
participant Model as HF Model
participant Wrap as torch_attention_hf_wrapper
participant Op as auto_deploy.torch_attention
rect rgba(220,235,255,0.35)
note over Patch,Reg: Patch application (new)
User->>Patch: apply()
Patch->>Reg: Reg["ad_unified_attn"] = Wrap
Patch->>te: patch te.export -> _export_with_unified_attn (sets config._attn_implementation)
end
User->>te: export(Model, ...)
te->>Model: run with _attn_implementation="ad_unified_attn"
Model->>Wrap: attention(q,k,v, attention_mask, **hf_kwargs)
note over Wrap: Reorder dims to "bsnd"<br/>Map kwargs via HF_ATTN_KWARGS_MAPPING
Wrap->>Op: torch_attention(q',k',v', mask, layout="bsnd", **mapped_kwargs)
Op-->>Wrap: output
Wrap-->>Model: output (restored layout)
Model-->>te: exported program
rect rgba(255,230,230,0.35)
note over Patch,Reg: Revert (on cleanup)
User->>Patch: revert()
Patch->>Reg: remove "ad_unified_attn"
Patch->>te: restore original te.export
end
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minutes Pre-merge checks and finishing touches❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (5)
tensorrt_llm/_torch/auto_deploy/custom_ops/torch_attention.py(1 hunks)tensorrt_llm/_torch/auto_deploy/export/library/unified_attn.py(1 hunks)tensorrt_llm/_torch/auto_deploy/models/patches/gptoss.py(0 hunks)tensorrt_llm/_torch/auto_deploy/transform/library/kvcache_transformers.py(2 hunks)tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_build_small_single.py(1 hunks)
💤 Files with no reviewable changes (1)
- tensorrt_llm/_torch/auto_deploy/models/patches/gptoss.py
🧰 Additional context used
📓 Path-based instructions (3)
**/*.{h,hpp,hh,hxx,cpp,cxx,cc,cu,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Use only spaces, no tabs; indent with 4 spaces.
Files:
tensorrt_llm/_torch/auto_deploy/transform/library/kvcache_transformers.pytests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_build_small_single.pytensorrt_llm/_torch/auto_deploy/custom_ops/torch_attention.pytensorrt_llm/_torch/auto_deploy/export/library/unified_attn.py
**/*.py
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.py: Python code must target Python 3.8+.
Indent Python code with 4 spaces; do not use tabs.
Maintain module namespace when importing; prefer 'from package.subpackage import foo' then 'foo.SomeClass()' instead of importing the class directly.
Python filenames should be snake_case (e.g., some_file.py).
Python classes use PascalCase names.
Functions and methods use snake_case names.
Local variables use snake_case; prefix 'k' for variables that start with a number (e.g., k_99th_percentile).
Global variables use upper SNAKE_CASE prefixed with 'G' (e.g., G_MY_GLOBAL).
Constants use upper SNAKE_CASE (e.g., MY_CONSTANT).
Avoid shadowing variables from an outer scope.
Initialize all externally visible members of a class in the constructor.
Prefer docstrings for interfaces that may be used outside a file; comments for in-function or file-local interfaces.
Use Google-style docstrings for classes and functions (Sphinx-parsable).
Document attributes and variables inline so they render under the class/function docstring.
Avoid reflection when a simpler, explicit approach suffices (e.g., avoid dict(**locals()) patterns).
In try/except, catch the most specific exceptions possible.
For duck-typing try/except, keep the try body minimal and use else for the main logic.
Files:
tensorrt_llm/_torch/auto_deploy/transform/library/kvcache_transformers.pytests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_build_small_single.pytensorrt_llm/_torch/auto_deploy/custom_ops/torch_attention.pytensorrt_llm/_torch/auto_deploy/export/library/unified_attn.py
**/*.{cpp,cxx,cc,h,hpp,hh,hxx,cu,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Prepend the NVIDIA Apache-2.0 copyright header with current year to the top of all source files (e.g., .cpp, .h, .cu, .py).
Files:
tensorrt_llm/_torch/auto_deploy/transform/library/kvcache_transformers.pytests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_build_small_single.pytensorrt_llm/_torch/auto_deploy/custom_ops/torch_attention.pytensorrt_llm/_torch/auto_deploy/export/library/unified_attn.py
🧬 Code graph analysis (2)
tensorrt_llm/_torch/auto_deploy/transform/library/kvcache_transformers.py (1)
tensorrt_llm/_torch/auto_deploy/custom_ops/torch_attention.py (1)
torch_attention(109-225)
tensorrt_llm/_torch/auto_deploy/export/library/unified_attn.py (2)
tensorrt_llm/_torch/auto_deploy/custom_ops/torch_attention.py (1)
torch_attention(109-225)tensorrt_llm/_torch/auto_deploy/export/interface.py (2)
BaseExportPatch(38-130)ExportPatchRegistry(166-213)
🪛 Ruff (0.13.3)
tensorrt_llm/_torch/auto_deploy/export/library/unified_attn.py
14-14: Unused function argument: self
(ARG001)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)
- GitHub Check: Pre-commit Check
📝 WalkthroughWalkthroughIntroduces a centralized HF-to-custom-op kwargs mapping, adds a unified export-time attention wrapper and patch that reroutes HuggingFace attention to a custom op, removes the prior GPT OSS patching module, updates a transformer-side KV cache transform to use the new mapping, and adjusts a unit test’s skip condition logic. Changes
Sequence Diagram(s)sequenceDiagram
autonumber
participant User as User
participant Export as te.export (patched)
participant Model as HF Model
participant AttnSel as ALL_ATTENTION_FUNCTIONS
participant Wrapper as torch_attention_hf_wrapper
participant Op as torch.ops.auto_deploy.torch_attention
User->>Export: export(Model)
rect rgba(220,235,245,0.4)
note right of Export: Patched export path when model.config._attn_implementation is set
Export->>AttnSel: register ad_unified_attn
Export->>Model: run export with patched attention
Model->>Wrapper: attention(query, key, value, mask, **hf_kwargs)
Wrapper->>Wrapper: reorder Q/K/V<br/>remap kwargs via HF_ATTN_KWARGS_MAPPING
Wrapper->>Op: torch_attention(q,k,v, mask, **mapped_kwargs)
Op-->>Wrapper: output
Wrapper-->>Model: output
end
Model-->>Export: graph
Export-->>User: exported artifact
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~25 minutes Pre-merge checks and finishing touches❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
There was a problem hiding this comment.
Actionable comments posted: 2
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (4)
tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_build_small_single.py (1)
1-1: Add NVIDIA Apache-2.0 header at file top.Per repo guidelines, prepend the copyright header.
As per coding guidelines
+# Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# http://www.apache.org/licenses/LICENSE-2.0 +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License.tensorrt_llm/_torch/auto_deploy/transform/library/kvcache_transformers.py (1)
1-1: Add NVIDIA Apache-2.0 header at file top.Please prepend the required header.
As per coding guidelines
+# Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# http://www.apache.org/licenses/LICENSE-2.0 +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License.tensorrt_llm/_torch/auto_deploy/custom_ops/torch_attention.py (2)
1-1: Add NVIDIA Apache-2.0 header at file top.Please prepend the required header.
As per coding guidelines
+# Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# http://www.apache.org/licenses/LICENSE-2.0 +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License.
18-31: Boolean attention mask conversion is incorrect (semantic bug).In PyTorch SDPA, boolean masks use True to indicate positions to mask out. Current conversion maps True→1.0, False→-inf, biasing allowed positions and inverting semantics.
Apply:
-def _convert_boolean_mask_to_float(attn_mask: torch.Tensor, dtype: torch.dtype) -> torch.Tensor: - """Convert boolean attention mask to floating point mask. - Args: - attn_mask: Boolean tensor where True allows attention, False blocks it - dtype: Target dtype for the output mask - Returns: - Floating point mask where True -> 1.0, False -> -inf - """ - if attn_mask.dtype == torch.bool: - float_mask = torch.zeros_like(attn_mask, dtype=dtype) - float_mask = float_mask.masked_fill(attn_mask, 1.0) # True -> 1.0 - float_mask = float_mask.masked_fill(~attn_mask, float("-inf")) # False -> -inf - return float_mask - return attn_mask +def _convert_boolean_mask_to_float(attn_mask: torch.Tensor, dtype: torch.dtype) -> torch.Tensor: + """Convert boolean attention mask (True = masked) to additive float mask. + Returns 0.0 for allowed positions and -inf for masked positions.""" + if attn_mask.dtype == torch.bool: + float_mask = torch.zeros_like(attn_mask, dtype=dtype) # allowed -> 0.0 + float_mask = float_mask.masked_fill(attn_mask, float("-inf")) # True (masked) -> -inf + return float_mask + return attn_maskThis matches F.scaled_dot_product_attention semantics.
🧹 Nitpick comments (3)
tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_build_small_single.py (1)
106-111: Skip logic looks correct for transformers mode.Condition now only triggers for transformers and specified models; good.
Optionally, tighten readability:
- if mode == "transformers" and ( - "DeepSeek-V3" in experiment_config["args"]["model"] - or "Phi-3-mini-4k-instruct" in experiment_config["args"]["model"] - or "NVIDIA-Nemotron-Nano-12B-v2" in experiment_config["args"]["model"] - ): + blocked = ("DeepSeek-V3", "Phi-3-mini-4k-instruct", "NVIDIA-Nemotron-Nano-12B-v2") + if mode == "transformers" and any(b in experiment_config["args"]["model"] for b in blocked):tensorrt_llm/_torch/auto_deploy/custom_ops/torch_attention.py (1)
94-105: HF kwargs mapping: LGTM; consider making immutable.Mapping covers common keys (dropout, scaling/scale, sinks, sliding_window, logit_cap, is_causal).
Optionally wrap with MappingProxyType for immutability:
+from types import MappingProxyType -HF_ATTN_KWARGS_MAPPING = { +HF_ATTN_KWARGS_MAPPING = MappingProxyType({ ... -} +})tensorrt_llm/_torch/auto_deploy/export/library/unified_attn.py (1)
13-21: Prefix unused module arg to satisfy linter and clarify intent.Rename
selfto_moduleto match HF signature and silence ARG001.-def torch_attention_hf_wrapper( - self: torch.nn.Module, +def torch_attention_hf_wrapper( + _module: torch.nn.Module,
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (5)
tensorrt_llm/_torch/auto_deploy/custom_ops/torch_attention.py(1 hunks)tensorrt_llm/_torch/auto_deploy/export/library/unified_attn.py(1 hunks)tensorrt_llm/_torch/auto_deploy/models/patches/gptoss.py(0 hunks)tensorrt_llm/_torch/auto_deploy/transform/library/kvcache_transformers.py(2 hunks)tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_build_small_single.py(1 hunks)
💤 Files with no reviewable changes (1)
- tensorrt_llm/_torch/auto_deploy/models/patches/gptoss.py
🧰 Additional context used
📓 Path-based instructions (3)
**/*.{h,hpp,hh,hxx,cpp,cxx,cc,cu,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Use only spaces, no tabs; indent with 4 spaces.
Files:
tensorrt_llm/_torch/auto_deploy/custom_ops/torch_attention.pytensorrt_llm/_torch/auto_deploy/export/library/unified_attn.pytensorrt_llm/_torch/auto_deploy/transform/library/kvcache_transformers.pytests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_build_small_single.py
**/*.py
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.py: Python code must target Python 3.8+.
Indent Python code with 4 spaces; do not use tabs.
Maintain module namespace when importing; prefer 'from package.subpackage import foo' then 'foo.SomeClass()' instead of importing the class directly.
Python filenames should be snake_case (e.g., some_file.py).
Python classes use PascalCase names.
Functions and methods use snake_case names.
Local variables use snake_case; prefix 'k' for variables that start with a number (e.g., k_99th_percentile).
Global variables use upper SNAKE_CASE prefixed with 'G' (e.g., G_MY_GLOBAL).
Constants use upper SNAKE_CASE (e.g., MY_CONSTANT).
Avoid shadowing variables from an outer scope.
Initialize all externally visible members of a class in the constructor.
Prefer docstrings for interfaces that may be used outside a file; comments for in-function or file-local interfaces.
Use Google-style docstrings for classes and functions (Sphinx-parsable).
Document attributes and variables inline so they render under the class/function docstring.
Avoid reflection when a simpler, explicit approach suffices (e.g., avoid dict(**locals()) patterns).
In try/except, catch the most specific exceptions possible.
For duck-typing try/except, keep the try body minimal and use else for the main logic.
Files:
tensorrt_llm/_torch/auto_deploy/custom_ops/torch_attention.pytensorrt_llm/_torch/auto_deploy/export/library/unified_attn.pytensorrt_llm/_torch/auto_deploy/transform/library/kvcache_transformers.pytests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_build_small_single.py
**/*.{cpp,cxx,cc,h,hpp,hh,hxx,cu,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Prepend the NVIDIA Apache-2.0 copyright header with current year to the top of all source files (e.g., .cpp, .h, .cu, .py).
Files:
tensorrt_llm/_torch/auto_deploy/custom_ops/torch_attention.pytensorrt_llm/_torch/auto_deploy/export/library/unified_attn.pytensorrt_llm/_torch/auto_deploy/transform/library/kvcache_transformers.pytests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_build_small_single.py
🧬 Code graph analysis (2)
tensorrt_llm/_torch/auto_deploy/export/library/unified_attn.py (2)
tensorrt_llm/_torch/auto_deploy/custom_ops/torch_attention.py (1)
torch_attention(109-225)tensorrt_llm/_torch/auto_deploy/export/interface.py (2)
BaseExportPatch(38-130)ExportPatchRegistry(166-213)
tensorrt_llm/_torch/auto_deploy/transform/library/kvcache_transformers.py (1)
tensorrt_llm/_torch/auto_deploy/custom_ops/torch_attention.py (1)
torch_attention(109-225)
🪛 Ruff (0.13.3)
tensorrt_llm/_torch/auto_deploy/export/library/unified_attn.py
14-14: Unused function argument: self
(ARG001)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (2)
- GitHub Check: Check PR Checklist Resolution
- GitHub Check: Pre-commit Check
🔇 Additional comments (1)
tensorrt_llm/_torch/auto_deploy/transform/library/kvcache_transformers.py (1)
14-15: Centralized kwargs mapping: LGTM.Importing and iterating HF_ATTN_KWARGS_MAPPING removes duplication and reduces drift; good alignment with export wrapper.
Ensure HF_ATTN_KWARGS_MAPPING stays in sync with any future HF attention_interface keys.
Also applies to: 43-46
lucaslie
left a comment
lucaslie
left a comment
There was a problem hiding this comment.
looks good. Just a few minor comments. Please slack me when those are resolved and I can kick off a dashboard run
h-guo18
force-pushed
the
haoguo/attn_export_patch
branch
from
1af2f4f to
3d2a36c
Compare
tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_build_small_single.py
Show resolved
Hide resolved
h-guo18
force-pushed
the
haoguo/attn_export_patch
branch
from
3d2a36c to
a280014
Compare
|
/bot run |
|
PR_Github #22243 [ run ] triggered by Bot. Commit: |
|
PR_Github #22243 [ run ] completed with state |
Signed-off-by: h-guo18 <67671475+h-guo18@users.noreply.github.com>
Signed-off-by: h-guo18 <67671475+h-guo18@users.noreply.github.com>
Signed-off-by: h-guo18 <67671475+h-guo18@users.noreply.github.com>
Signed-off-by: h-guo18 <67671475+h-guo18@users.noreply.github.com>
h-guo18
force-pushed
the
haoguo/attn_export_patch
branch
from
a280014 to
d42f58b
Compare
|
/bot run |
|
PR_Github #22317 [ run ] triggered by Bot. Commit: |
|
PR_Github #22317 [ run ] completed with state |
3 participants
Summary by CodeRabbit
New Features
Refactor
Description
Added a export patch to replace
hf.attention_interfacedirectly to the unifiedautodeploy::torch_attention. Map keywords likesliding_window,scaling,sinkduring patching.This allow us to support the attention keywords above for different models generally.
Test Coverage
Build_and_run
Tested on
unsloth/gpt-oss-20b-BF16andllama-3.1-8B, got identical output with upstream/main andnum_matches=0in match_eager_attn and match_grouped_attn transform, indicating attention matching is bypassed:python examples/auto_deploy/build_and_run_ad.py --model unsloth/gpt-oss-20b-BF16 --args.world_size 0 --args.runtime "demollm" --args.attn_backend torch --args.compile_backend torch-simpleNOTE: Output is strange but this is identical to result on
upstream/main.python examples/auto_deploy/build_and_run_ad.py --model meta-llama/Llama-3.1-8B-Instruct --args.world_size 0 --args.runtime "demollm"Unit test
Only failing same cases as upstream/main:
PR Checklist
Please review the following before submitting your PR:
PR description clearly explains what and why. If using CodeRabbit's summary, please make sure it makes sense.
PR Follows TRT-LLM CODING GUIDELINES to the best of your knowledge.
Test cases are provided for new code paths (see test instructions)
Any new dependencies have been scanned for license and vulnerabilities
CODEOWNERS updated if ownership changes
Documentation updated as needed
The reviewers assigned automatically/manually are appropriate for the PR.
Please check this after reviewing the above items as appropriate for this PR.
GitHub Bot Help
/bot [-h] ['run', 'kill', 'skip', 'reuse-pipeline'] ...Provide a user friendly way for developers to interact with a Jenkins server.
Run
/bot [-h|--help]to print this help message.See details below for each supported subcommand.
Details
run [--reuse-test (optional)pipeline-id --disable-fail-fast --skip-test --stage-list "A10-PyTorch-1, xxx" --gpu-type "A30, H100_PCIe" --test-backend "pytorch, cpp" --add-multi-gpu-test --only-multi-gpu-test --disable-multi-gpu-test --post-merge --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" --detailed-log --debug(experimental)]Launch build/test pipelines. All previously running jobs will be killed.
--reuse-test (optional)pipeline-id(OPTIONAL) : Allow the new pipeline to reuse build artifacts and skip successful test stages from a specified pipeline or the last pipeline if no pipeline-id is indicated. If the Git commit ID has changed, this option will be always ignored. The DEFAULT behavior of the bot is to reuse build artifacts and successful test results from the last pipeline.--disable-reuse-test(OPTIONAL) : Explicitly prevent the pipeline from reusing build artifacts and skipping successful test stages from a previous pipeline. Ensure that all builds and tests are run regardless of previous successes.--disable-fail-fast(OPTIONAL) : Disable fail fast on build/tests/infra failures.--skip-test(OPTIONAL) : Skip all test stages, but still run build stages, package stages and sanity check stages. Note: Does NOT update GitHub check status.--stage-list "A10-PyTorch-1, xxx"(OPTIONAL) : Only run the specified test stages. Examples: "A10-PyTorch-1, xxx". Note: Does NOT update GitHub check status.--gpu-type "A30, H100_PCIe"(OPTIONAL) : Only run the test stages on the specified GPU types. Examples: "A30, H100_PCIe". Note: Does NOT update GitHub check status.--test-backend "pytorch, cpp"(OPTIONAL) : Skip test stages which don't match the specified backends. Only support [pytorch, cpp, tensorrt, triton]. Examples: "pytorch, cpp" (does not run test stages with tensorrt or triton backend). Note: Does NOT update GitHub pipeline status.--only-multi-gpu-test(OPTIONAL) : Only run the multi-GPU tests. Note: Does NOT update GitHub check status.--disable-multi-gpu-test(OPTIONAL) : Disable the multi-GPU tests. Note: Does NOT update GitHub check status.--add-multi-gpu-test(OPTIONAL) : Force run the multi-GPU tests in addition to running L0 pre-merge pipeline.--post-merge(OPTIONAL) : Run the L0 post-merge pipeline instead of the ordinary L0 pre-merge pipeline.--extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx"(OPTIONAL) : Run the ordinary L0 pre-merge pipeline and specified test stages. Examples: --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx".--detailed-log(OPTIONAL) : Enable flushing out all logs to the Jenkins console. This will significantly increase the log volume and may slow down the job.--debug(OPTIONAL) : Experimental feature. Enable access to the CI container for debugging purpose. Note: Specify exactly one stage in thestage-listparameter to access the appropriate container environment. Note: Does NOT update GitHub check status.For guidance on mapping tests to stage names, see
docs/source/reference/ci-overview.mdand the
scripts/test_to_stage_mapping.pyhelper.kill
killKill all running builds associated with pull request.
skip
skip --comment COMMENTSkip testing for latest commit on pull request.
--comment "Reason for skipping build/test"is required. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.reuse-pipeline
reuse-pipelineReuse a previous pipeline to validate current commit. This action will also kill all currently running builds associated with the pull request. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.