[TRTLLM-9752][fix] WAR: Disable PDL for quant kernels to fix accuracy issues #10285
Conversation
📝 WalkthroughWalkthroughProgrammatic stream serialization is disabled across three CUDA quantization code paths (FP4 quantization, MXFP8 quantization, and per-token FP4 global-scale computation) by replacing dynamic environment variable checks with hardcoded false values in CUDA launch configurations. Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~8 minutes Pre-merge checks and finishing touches❌ Failed checks (2 warnings)
✅ Passed checks (1 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: 0
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
cpp/tensorrt_llm/kernels/quantization.cu (1)
20-20: Remove unusedenvUtils.hinclude on line 20.After removing the PDL environment variable check (
getEnvEnablePDL()), theenvUtils.hinclude is no longer referenced anywhere in the file and has no transitive dependencies through other headers. Remove this include to clean up unused dependencies.
🧹 Nitpick comments (1)
cpp/tensorrt_llm/kernels/quantization.cu (1)
173-184: PDL disabled for FP4 quantization - change looks correct.The programmatic stream serialization is now hardcoded to false. The implementation is correct and consistent with the other changes in this PR.
Consider adding a comment explaining why PDL is disabled
While not strictly necessary, adding a brief comment would help future maintainers understand the rationale:
cudaLaunchAttribute attrs[1]; attrs[0].id = cudaLaunchAttributeProgrammaticStreamSerialization; +// PDL disabled for quant kernels (TRTLLM-9752) attrs[0].val.programmaticStreamSerializationAllowed = false;
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
cpp/tensorrt_llm/kernels/quantization.cu
🧰 Additional context used
📓 Path-based instructions (2)
**/*.{cpp,h,cu,cuh}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.{cpp,h,cu,cuh}: Closing braces of namespaces should have a comment saying the namespace it closes:} // namespace foo
Preferconstorconstexprvariables over#definewhenever possible, as the latter are not visible to the compiler
A variable that is not modified after its initialization should be declared asconst
For naming of constants in C++, follow the naming section conventions
Except0(only used in comparison for checking signness/existence/emptiness) andnullptr,true,false, all other literals should only be used for variable initialization in C++
Use the Allman indentation style in C++
Put the semicolon for an emptyfororwhileloop in a new line in C++
The statement forming the body of aswitch,while,do .. whileorforstatement shall be a compound statement (use brace-delimited statements) in C++
If and else should always be followed by brace-delimited statements, even if empty or a single statement in C++
C++ filenames should use camel case with first letter lowercase:thisIsASubDirandthisIsAFilename.cpp
All files involved in the compilation of a compilation target (.exe/.so) must have filenames that are case-insensitive unique in C++
All types (including class names) in C++ should use camel case with uppercase first letter:FooBarClass
Local variables, methods and namespaces in C++ should use camel case with first letter lowercase:localFooBar
Non-magic-number global variables that are non-static and not defined in anonymous namespace in C++ should use camel case prefixed by a lower case 'g':gDontUseGlobalFoos
Non-magic-number global variables that are static or defined in an anonymous namespace in C++ should use camel case prefixed by a lower case 's':sMutableStaticGlobal
Locally visible static variables in C++ should use camel case with lowercase prefix 's' as the first letter:static std::once_flag sFlag;
Public, private and protected class member variables in C++ should use camel case prefi...
Files:
cpp/tensorrt_llm/kernels/quantization.cu
**/*.{cpp,h,cu,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
All TensorRT-LLM Open Source Software code should contain an NVIDIA copyright header that includes the year of its latest meaningful modification
Files:
cpp/tensorrt_llm/kernels/quantization.cu
🧠 Learnings (10)
📓 Common learnings
Learnt from: venkywonka
Repo: NVIDIA/TensorRT-LLM PR: 6029
File: .github/pull_request_template.md:45-53
Timestamp: 2025-08-27T17:50:13.264Z
Learning: For PR templates in TensorRT-LLM, avoid suggesting changes that would increase developer overhead, such as converting plain bullets to mandatory checkboxes. The team prefers guidance-style bullets that don't require explicit interaction to reduce friction in the PR creation process.
📚 Learning: 2025-09-23T15:01:00.070Z
Learnt from: nv-lschneider
Repo: NVIDIA/TensorRT-LLM PR: 7910
File: cpp/tensorrt_llm/kernels/nccl_device/config.cu:15-17
Timestamp: 2025-09-23T15:01:00.070Z
Learning: In TensorRT-LLM NCCL device kernels, the <sstream> header is not needed as an explicit include in config.cu because it's provided transitively through other headers. Local compilation testing confirms this works without the explicit include.
Applied to files:
cpp/tensorrt_llm/kernels/quantization.cu
📚 Learning: 2025-09-23T15:01:00.070Z
Learnt from: nv-lschneider
Repo: NVIDIA/TensorRT-LLM PR: 7910
File: cpp/tensorrt_llm/kernels/nccl_device/config.cu:15-17
Timestamp: 2025-09-23T15:01:00.070Z
Learning: In TensorRT-LLM NCCL device kernels (cpp/tensorrt_llm/kernels/nccl_device/config.cu), std::ostringstream is used but <sstream> doesn't need to be explicitly included because it's provided transitively through other headers like tensorrt_llm/common/cudaUtils.h or config.h. Local compilation testing confirms this works without the explicit include.
Applied to files:
cpp/tensorrt_llm/kernels/quantization.cu
📚 Learning: 2025-09-23T15:13:48.819Z
Learnt from: nv-lschneider
Repo: NVIDIA/TensorRT-LLM PR: 7910
File: cpp/tensorrt_llm/kernels/nccl_device/multimem.h:20-30
Timestamp: 2025-09-23T15:13:48.819Z
Learning: TRT-LLM targets modern CUDA toolkits that support FP8 datatypes, so cuda_fp8.h can be included unconditionally without version guards in TRT-LLM code.
Applied to files:
cpp/tensorrt_llm/kernels/quantization.cu
📚 Learning: 2025-11-14T11:22:03.729Z
Learnt from: nzmora-nvidia
Repo: NVIDIA/TensorRT-LLM PR: 9163
File: tensorrt_llm/_torch/auto_deploy/custom_ops/quant.py:107-113
Timestamp: 2025-11-14T11:22:03.729Z
Learning: In TensorRT-LLM AutoDeploy custom ops, when adding hardware capability checks to select between kernel implementations (e.g., cuBLAS vs. CUDA kernel), use descriptive variable names that identify the specific GPU architectures or families being targeted (e.g., `is_blackwell_geforce_or_ada`) rather than generic names like `enable_cuda_core`. This makes it clear that the code is selecting an implementation path based on hardware capabilities, not enabling/disabling hardware features.
Applied to files:
cpp/tensorrt_llm/kernels/quantization.cu
📚 Learning: 2025-12-19T06:31:54.973Z
Learnt from: nvyocox
Repo: NVIDIA/TensorRT-LLM PR: 10117
File: tensorrt_llm/_torch/auto_deploy/transform/library/fuse_rope_attention.py:336-339
Timestamp: 2025-12-19T06:31:54.973Z
Learning: In tensorrt_llm/_torch/auto_deploy/transform/library/fuse_rope_attention.py, the cast to torch.float16 for qkv_node before creating the AttentionPlugin is intentional and required because DriveOS LLM expects float16 dtype specifically. This should not be changed to preserve original dtype or made configurable for bfloat16 models in the DriveOS LLM ONNX export path.
Applied to files:
cpp/tensorrt_llm/kernels/quantization.cu
📚 Learning: 2025-08-17T15:07:01.420Z
Learnt from: amitz-nv
Repo: NVIDIA/TensorRT-LLM PR: 6968
File: cpp/tensorrt_llm/thop/loraOp.cpp:133-141
Timestamp: 2025-08-17T15:07:01.420Z
Learning: In TensorRT-LLM's LoRA implementation, the LoraImpl::run() method handles setStream() internally in _runGemm() (line 51 in lora.cpp), along with setWorkspace(). The stream parameter flows from loraOp.cpp through LoraImpl::run() to _runGemm() where setStream() is called appropriately. Adding setStream() in loraOp.cpp would be redundant and goes against the intended architectural design.
Applied to files:
cpp/tensorrt_llm/kernels/quantization.cu
📚 Learning: 2025-08-14T15:43:23.107Z
Learnt from: MatthiasKohl
Repo: NVIDIA/TensorRT-LLM PR: 6904
File: tensorrt_llm/_torch/attention_backend/trtllm.py:259-262
Timestamp: 2025-08-14T15:43:23.107Z
Learning: In TensorRT-LLM's attention backend, tensor parameters in the plan() method are assigned directly without validation (dtype, device, contiguity checks). This maintains consistency across all tensor inputs and follows the pattern of trusting callers to provide correctly formatted tensors.
Applied to files:
cpp/tensorrt_llm/kernels/quantization.cu
📚 Learning: 2025-08-26T09:37:10.463Z
Learnt from: jiaganc
Repo: NVIDIA/TensorRT-LLM PR: 7031
File: tensorrt_llm/bench/dataclasses/configuration.py:90-104
Timestamp: 2025-08-26T09:37:10.463Z
Learning: In TensorRT-LLM, the `get_pytorch_perf_config()` method returns `self.pytorch_config` which can contain default `cuda_graph_config` values, so `llm_args` may already have this config before the extra options processing.
Applied to files:
cpp/tensorrt_llm/kernels/quantization.cu
📚 Learning: 2025-08-17T15:07:01.420Z
Learnt from: amitz-nv
Repo: NVIDIA/TensorRT-LLM PR: 6968
File: cpp/tensorrt_llm/thop/loraOp.cpp:133-141
Timestamp: 2025-08-17T15:07:01.420Z
Learning: In TensorRT-LLM's LoRA implementation, the LoraImpl::run() method handles setStream() internally in _runGemm(), along with setWorkspace(). Both stream and workspace are passed as arguments to run(), so there's no need to call setStream() explicitly in loraOp.cpp - this avoids redundancy and follows the intended architectural separation.
Applied to files:
cpp/tensorrt_llm/kernels/quantization.cu
⏰ 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
🔇 Additional comments (2)
cpp/tensorrt_llm/kernels/quantization.cu (2)
208-220: PDL disabled for MXFP8 quantization - change looks correct.The implementation is consistent with the FP4 quantization path and correctly disables programmatic stream serialization.
383-394: PDL disabled for per-token global scale computation - change looks correct.The implementation is consistent with the other two changes and correctly disables programmatic stream serialization.
Signed-off-by: Bo Deng <[email protected]>
bo-nv
force-pushed
the
user/bo/1.2.0rc2-fix-accuracy
branch
from
5d778a6 to
0163e1d
Compare
bo-nv
changed the title
|
/bot run |
|
PR_Github #29875 [ run ] triggered by Bot. Commit: |
bo-nv
changed the title
|
/bot run |
|
PR_Github #29875 [ run ] completed with state
|
|
PR_Github #29905 [ run ] triggered by Bot. Commit: |
|
PR_Github #29905 [ run ] completed with state |
3 participants
This is just a temporary WAR for the accuracy issue mentioned in TRTLLM-9752. We will submit another PR to fix it once we find the root cause.
Summary by CodeRabbit
✏️ Tip: You can customize this high-level summary in your review settings.
Description
Test Coverage
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.