[None][fix] Skip Topk if 0 #6934
Conversation
📝 WalkthroughWalkthroughGuarded top-k filtering in top_k_top_p_sampling_batch so when top_k <= 0 the top-k path is skipped. The min_values computation and -inf masking were moved inside the top_k > 0 branch; top-p sorting, masking, softmax, and sampling remain unchanged. Changes
Sequence Diagram(s)sequenceDiagram
participant Caller
participant Sampler as top_k_top_p_sampling_batch
Caller->>Sampler: logits, top_k, top_p, temperature
alt top_k > 0
Sampler->>Sampler: compute top-k (min_values)
Sampler->>Sampler: mask logits outside top-k (-inf)
else top_k <= 0
Note right of Sampler: Skip top-k masking
end
Sampler->>Sampler: sort logits for top-p
Sampler->>Sampler: mask probs outside top_p
Sampler->>Sampler: softmax, sample token
Sampler-->>Caller: sampled token
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~12 minutes Suggested reviewers
Tip 🔌 Remote MCP (Model Context Protocol) integration is now available!Pro plan users can now connect to remote MCP servers from the Integrations page. Connect with popular remote MCPs such as Notion and Linear to add more context to your reviews and chats. 📜 Recent review detailsConfiguration used: .coderabbit.yaml 💡 Knowledge Base configuration:
You can enable these sources in your CodeRabbit configuration. 📒 Files selected for processing (1)
🚧 Files skipped from review as they are similar to previous changes (1)
✨ Finishing Touches
🧪 Generate unit tests
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. 🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. CodeRabbit Commands (Invoked using PR/Issue comments)Type Other keywords and placeholders
Status, Documentation and Community
|
![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
🔭 Outside diff range comments (4)
tensorrt_llm/_torch/pyexecutor/sampler.py (4)
168-174: Fix stale-dimension assert — currently fails for 1D input after unsqueeze
logits_dimis captured before the potentialunsqueeze. If 1D is passed, you unsqueeze to 2D but still assert on the stalelogits_dim == 2, which is false (1 == 2) and raises. Assert onlogits.dim()instead.- assert logits_dim == 2, "logits should be 2D: [batch_size, vocab_size]" + assert logits.dim() == 2, "logits should be 2D: [batch_size, vocab_size]"
130-134: Same stale-dimension assert bug in top‑p pathThe same issue exists here; assert against
logits.dim()after the optionalunsqueeze.- assert logits_dim == 2, "logits should be 2D: [batch_size, vocab_size]" + assert logits.dim() == 2, "logits should be 2D: [batch_size, vocab_size]"
243-249: Type hint fix:Strategyexcludes thetop_k_top_pvariant
sample()pattern-matches a("top_k_top_p", ...)case, butStrategyomitsTopKTopP. Include it to keep type hints consistent and avoid static analysis noise.-Strategy = TopK | TopP | Greedy +Strategy = TopK | TopP | TopKTopP | Greedy
1-1: Add NVIDIA copyright headerPer coding guidelines, prepend the current-year NVIDIA copyright header.
+# Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. + from abc import ABC, abstractmethod
🧹 Nitpick comments (2)
tensorrt_llm/_torch/pyexecutor/sampler.py (2)
181-183: Usemasked_fillfor clarity and a bit less overheadSlightly cleaner and avoids creating a full
-inftensor viatorch.where.- logits = torch.where(logits < min_values, - torch.full_like(logits, float('-inf')), logits) + logits = logits.masked_fill(logits < min_values, float('-inf'))
193-194: Nit: use booleanFalseinstead of0for readability
sorted_indices_to_removeis boolean; preferFalseover0.- sorted_indices_to_remove[:, 0] = 0 + sorted_indices_to_remove[:, 0] = False
📜 Review details
Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
💡 Knowledge Base configuration:
- MCP integration is disabled by default for public repositories
- Jira integration is disabled by default for public repositories
- Linear integration is disabled by default for public repositories
You can enable these sources in your CodeRabbit configuration.
📒 Files selected for processing (1)
tensorrt_llm/_torch/pyexecutor/sampler.py(1 hunks)
🧰 Additional context used
📓 Path-based instructions (2)
**/*.py
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
**/*.py: Python code must target Python 3.8+
Python indentation: 4 spaces, no tabs
Maintain module namespace in imports (from package.subpackage import foo; then use foo.SomeClass())
Python file names use snake_case
Python class names use PascalCase
Python functions/methods and local variables use snake_case; variables starting with a number get k_ prefix (e.g., k_99th_percentile)
Global variables use G_ prefixed UPPER_SNAKE_CASE (e.g., G_MY_GLOBAL)
Constants use UPPER_SNAKE_CASE in Python
Avoid shadowing variables from outer scopes in Python
Initialize all externally visible members of a Python class in init
Prefer docstrings for interfaces used outside a file; comments for local code
Use Google-style docstrings for classes and functions (Sphinx-parsable)
Document attributes/variables inline with short docstrings
Avoid reflection when simple alternatives exist (e.g., prefer explicit parameters over dict(**locals()))
In try/except, catch the narrowest exceptions possible
For duck-typing with try/except, keep try body minimal and put logic in else
Files:
tensorrt_llm/_torch/pyexecutor/sampler.py
**/*.{cpp,cxx,cc,cu,h,hpp,hxx,hh,cuh,py}
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
Prepend NVIDIA copyright header (current year) to all source files
Files:
tensorrt_llm/_torch/pyexecutor/sampler.py
🔇 Additional comments (1)
tensorrt_llm/_torch/pyexecutor/sampler.py (1)
176-183: Top‑k guarded bytop_k > 0— matches PR intentGuarding the top‑k masking behind
if top_k > 0:correctly enables “skip top‑k when 0 or negative.” This aligns with the PR objective. Nice and minimal change.
|
/bot run |
|
PR_Github #15463 [ run ] triggered by Bot |
|
PR_Github #15463 [ run ] completed with state |
Signed-off-by: Izzy Putterman <[email protected]>
Signed-off-by: Izzy Putterman <[email protected]>
hlu1
force-pushed
the
iputterman/topktopp-fix
branch
from
74ab679 to
8eba607
Compare
|
/bot reuse-pipeline |
|
PR_Github #15482 [ reuse-pipeline ] triggered by Bot |
|
@schetlur-nv Yeah in the OAI protocol it defaults to 0: https://github.com/NVIDIA/TensorRT-LLM/blob/main/tensorrt_llm/serve/openai_protocol.py#L217 |
|
thanks, am less worried about that prob, but please fix the other place where top_k sampling is done. Seems to have the same problem if I read it right, |
Signed-off-by: Izzy Putterman <[email protected]>
Head branch was pushed to by a user without write access
|
PR_Github #15482 [ reuse-pipeline ] completed with state |
|
/bot run |
|
PR_Github #15489 [ run ] triggered by Bot |
|
PR_Github #15489 [ run ] completed with state |
|
/bot run --disable-fail-fast |
|
PR_Github #15498 [ run ] triggered by Bot |
|
PR_Github #15498 [ run ] completed with state |
|
/bot skip --comment "#15463 is good. #15498 has one timeout very late in the pipeline. Should be safe." |
|
PR_Github #15506 [ skip ] triggered by Bot |
|
PR_Github #15506 [ skip ] completed with state |
Signed-off-by: Izzy Putterman <[email protected]>
Signed-off-by: Izzy Putterman <[email protected]> Signed-off-by: Wangshanshan <[email protected]>
Signed-off-by: Izzy Putterman <[email protected]> Signed-off-by: Wangshanshan <[email protected]>
Signed-off-by: Izzy Putterman <[email protected]>
Signed-off-by: Izzy Putterman <[email protected]> Signed-off-by: Wangshanshan <[email protected]>
Signed-off-by: Izzy Putterman <[email protected]>
Signed-off-by: Izzy Putterman <[email protected]> Signed-off-by: Wangshanshan <[email protected]>
5 participants
Summary by CodeRabbit
New Features
Bug Fixes
Performance
Documentation
Description
Test Coverage
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.
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.