[https://nvbugs/5599086][fix] Fix FP8 Linear module for spark #8707
Conversation
📝 WalkthroughWalkthroughThe FP8QDQLinearMethod class in Changes
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 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
🧹 Nitpick comments (2)
tensorrt_llm/_torch/modules/linear.py (2)
452-462: Simplify device handling logic and avoid multiple Parameter reassignments.The current implementation has a redundant conditional check and creates new Parameter objects twice during device transfers. This can be simplified:
- The second check
if original_device != module.weight_scale.device:(line 460) will only be true if the first condition at line 455 was true, making it redundant.- Creating new Parameter objects repeatedly can affect object identity and may have unintended side effects.
Consider this cleaner approach:
- original_device = module.weight_scale.device - # module.weight_scale and fused_weight must be on the same device for division operation - # Reset the device of module.weight_scale to the original device if changed + # Ensure module.weight_scale and fused_weight are on the same device for division if module.weight_scale.device != fused_weight.device: - module.weight_scale = Parameter( - module.weight_scale.data.to(fused_weight.device)) - fused_weight = (fused_weight / module.weight_scale).to( - torch.float8_e4m3fn) - if original_device != module.weight_scale.device: - module.weight_scale = Parameter( - module.weight_scale.data.to(original_device)) + weight_scale_for_div = module.weight_scale.to(fused_weight.device) + else: + weight_scale_for_div = module.weight_scale + fused_weight = (fused_weight / weight_scale_for_div).to(torch.float8_e4m3fn)This approach:
- Eliminates the redundant second conditional check
- Avoids mutating the Parameter object, preserving object identity
- Achieves the same result with clearer logic
498-508: Simplify device handling logic and avoid multiple Parameter reassignments.The implementation here has the same issues as in
load_weights_fused_qkv_linear: redundant conditional check and multiple Parameter object creations.Apply the same simplification as suggested for the QKV method:
- original_device = module.weight_scale.device - # module.weight_scale and fused_weight must be on the same device for division operation - # Reset the device of module.weight_scale to the original device if changed if module.weight_scale.device != fused_weight.device: - module.weight_scale = Parameter( - module.weight_scale.data.to(fused_weight.device)) - fused_weight = (fused_weight / module.weight_scale).to( - torch.float8_e4m3fn) - if original_device != module.weight_scale.device: - module.weight_scale = Parameter( - module.weight_scale.data.to(original_device)) + weight_scale_for_div = module.weight_scale.to(fused_weight.device) + else: + weight_scale_for_div = module.weight_scale + fused_weight = (fused_weight / weight_scale_for_div).to(torch.float8_e4m3fn)
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
tensorrt_llm/_torch/modules/linear.py(2 hunks)
🧰 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/modules/linear.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/modules/linear.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/modules/linear.py
⏰ 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 (1)
tensorrt_llm/_torch/modules/linear.py (1)
452-462: Device transfer fix is correct and necessary.Verification confirms the fix is required for correctness:
Root cause of device mismatch: On integrated GPUs,
load_weight_shardskips device transfers (lines 73-80) and weights stay on CPU, whileweight_scaleloaded from checkpoint ends up on the default GPU device.Fix is necessary: The division operation at line 458/504 requires both
fused_weightandmodule.weight_scaleon the same device. The temporary device swap enables this without conflicting with the integrated GPU optimization.Symmetric implementation: Both
load_weights_fused_qkv_linear(lines 452-462) andload_weights_fused_gate_up_linear(lines 498-508) have identical fixes, which is correct.The fix properly handles the device mismatch that results from the memory optimization on integrated GPUs and causes no performance concern.
|
/bot run |
|
PR_Github #22711 [ run ] triggered by Bot. Commit: |
|
PR_Github #22711 [ run ] completed with state |
|
/bot run |
|
PR_Github #22796 [ run ] triggered by Bot. Commit: |
|
PR_Github #22796 [ run ] completed with state |
|
/bot run |
|
PR_Github #22906 [ run ] triggered by Bot. Commit: |
|
PR_Github #22906 [ run ] completed with state |
Signed-off-by: Simeng Liu <[email protected]>
Signed-off-by: Simeng Liu <[email protected]>
SimengLiu-nv
force-pushed
the
fix-spark-fp8
branch
from
422ae7c to
b31d0ad
Compare
|
/bot reuse-pipeline |
|
PR_Github #22915 [ reuse-pipeline ] triggered by Bot. Commit: |
|
PR_Github #22915 [ reuse-pipeline ] completed with state |
…#8707) Signed-off-by: Simeng Liu <[email protected]>
4 participants
Summary by CodeRabbit
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
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.
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.