[TRTLLM-10030][perf] avoid syncs in beam search + other improvements#11349
[TRTLLM-10030][perf] avoid syncs in beam search + other improvements#11349ixlmar merged 2 commits intoNVIDIA:mainfrom
Conversation
|
/bot run --disable-multi-gpu-test |
|
/bot run |
📝 WalkthroughWalkthroughThis PR introduces lazy beam history construction in the torch sampler using a builder pattern, where Changes
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~22 minutes 🚥 Pre-merge checks | ✅ 2 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ 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: 0
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
tensorrt_llm/_torch/pyexecutor/sampler.py (1)
1745-1756:⚠️ Potential issue | 🟠 MajorFix slicing logic to handle
preallocate_extra_steps=0correctly.When
preallocate_extra_steps=0, the slicelogprobs_tensor_full[:, :-0, :]produces an empty tensor (equivalent to[:, :0, :]), causing the filling loop to be skipped and the function to return uninitialized tensors. Although the current sole caller passespreallocate_extra_steps=1explicitly, the default value of0in the function signature creates a correctness trap for future use.Apply the proposed fix to check if
preallocate_extra_steps > 0before slicing:Fix
- logprobs_tensor = logprobs_tensor_full[:, :-preallocate_extra_steps, :] - logprobs_indices_tensor = logprobs_indices_tensor_full[:, :-preallocate_extra_steps, :] + if preallocate_extra_steps > 0: + logprobs_tensor = logprobs_tensor_full[:, :-preallocate_extra_steps, :] + logprobs_indices_tensor = logprobs_indices_tensor_full[:, :-preallocate_extra_steps, :] + else: + logprobs_tensor = logprobs_tensor_full + logprobs_indices_tensor = logprobs_indices_tensor_full
🧹 Nitpick comments (1)
tensorrt_llm/_torch/pyexecutor/sampler.py (1)
1280-1280: Static analysis:zip()withoutstrict=parameter.Ruff flags this
zip(token_list[beam_idx], logprobs_list[beam_idx])as missing an explicitstrict=parameter. Since both lists are derived from tensors of the same shape,strict=Truewould be appropriate and would catch shape mismatches early.Proposed fix
- for topk_token, topk_logprob in zip(token_list[beam_idx], logprobs_list[beam_idx]): + for topk_token, topk_logprob in zip(token_list[beam_idx], logprobs_list[beam_idx], strict=True):
|
PR_Github #35114 [ run ] triggered by Bot. Commit: |
|
PR_Github #35115 [ run ] triggered by Bot. Commit: |
|
PR_Github #35114 [ run ] completed with state |
|
/bot run --disable-fail-fast |
|
PR_Github #35139 [ run ] triggered by Bot. Commit: |
|
PR_Github #35115 [ run ] completed with state |
|
PR_Github #35139 [ run ] completed with state
|
Signed-off-by: ixlmar <206748156+ixlmar@users.noreply.github.com>
ixlmar
force-pushed
the
perf/beam-search-3
branch
from
077b567 to
918164e
Compare
|
In benchmarks using a 1B parameter model and short output sequence lengths, the performance advantage of |
Signed-off-by: ixlmar <206748156+ixlmar@users.noreply.github.com>
|
/bot run --disable-fail-fast |
|
PR_Github #35318 [ run ] triggered by Bot. Commit: |
|
/bot run --disable-fail-fast |
|
PR_Github #35320 [ run ] triggered by Bot. Commit: |
|
/bot run --disable-fail-fast |
|
PR_Github #35320 [ run ] completed with state
|
|
PR_Github #35342 [ run ] triggered by Bot. Commit: |
|
PR_Github #35342 [ run ] completed with state |
…VIDIA#11349) Signed-off-by: ixlmar <206748156+ixlmar@users.noreply.github.com> Signed-off-by: Ahmet Inci <ainci@nvidia.com>
3 participants
Description
Note: Commits up to and including 15087b5 are shared with #11341.
Details:
update_requests(avoids sync insample_async)update_requests(avoids sync)_convert_logprobs_tensor_to_listTest Coverage
n/a
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
Update tava architecture diagram if there is a significant design change in PR.
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.
Summary by CodeRabbit
Release Notes