[#11455][fix] Fallback to triton_ssm for nvfp4 quantization#11456
[#11455][fix] Fallback to triton_ssm for nvfp4 quantization#11456galagam merged 1 commit intoNVIDIA:mainfrom
Conversation
📝 WalkthroughWalkthroughThe changes add a new Changes
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 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)
Tip Issue Planner is now in beta. Read the docs and try it out! Share your feedback on Discord. 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: 2
🤖 Fix all issues with AI agents
In `@tensorrt_llm/_torch/auto_deploy/transform/library/ssm_cache.py`:
- Line 1: Add the required NVIDIA copyright header with the Apache License 2.0
at the top of this module (replace or precede the existing module docstring """A
set of transforms to handle SSM cache transforms."""), using the current year
2026 and the standard NVIDIA/Apache-2.0 boilerplate; ensure the header includes
the copyright owner (NVIDIA CORPORATION) and the full Apache-2.0 notice so the
file header meets the project's "year of latest meaningful modification"
requirement.
- Around line 25-26: The assignment to is_nvfp4 calls .upper() on
qcfg.get("quant_algo") which can be None; update the logic around
factory.get_quant_config() and the is_nvfp4 computation to guard against None by
coercing the value to a string or using a fallback (e.g., use
(qcfg.get("quant_algo") or "") or str(qcfg.get("quant_algo") or "") before
calling .upper()) so that None does not trigger an AttributeError when computing
is_nvfp4.
🧹 Nitpick comments (1)
tensorrt_llm/_torch/auto_deploy/transform/library/ssm_cache.py (1)
27-33: Persistent mutation ofself.config.backend— confirm this is safe.Line 32 permanently mutates
self.config.backend. If thisSSMCacheTransforminstance is ever reused (e.g., applied to multiple models), the fallback totriton_ssmwill persist silently without the warning being logged again (since the condition on line 27 won't match). If transforms are single-use per pipeline run, this is fine.
|
/bot run |
|
PR_Github #35664 [ run ] triggered by Bot. Commit: |
|
PR_Github #35664 [ run ] completed with state
|
|
/bot run --reuse-test |
|
PR_Github #35711 [ run ] triggered by Bot. Commit: |
|
PR_Github #35711 [ run ] completed with state
|
nvfp4 + flashinfer_ssm mamba backend is incompatible. Will be enabled as part of SuperV3 nvfp4 enablement. For now, falls back to triton_ssm Signed-off-by: Gal Hubara Agam <96368689+galagam@users.noreply.github.com>
galagam
force-pushed
the
gagam/war_for_fi_ssm_nvfp4
branch
from
9b3ab4f to
19390a0
Compare
|
/bot run |
|
/bot run --stage-list "DGX_B200-4_GPUs-AutoDeploy-1, DGX_H100-4_GPUs-AutoDeploy-1" --disable-fail-fast |
|
PR_Github #35737 [ run ] triggered by Bot. Commit: |
|
PR_Github #35739 [ run ] triggered by Bot. Commit: |
|
PR_Github #35739 [ run ] completed with state |
|
/bot run |
|
PR_Github #35775 [ run ] triggered by Bot. Commit: |
|
PR_Github #35775 [ run ] completed with state |
…IDIA#11456) Signed-off-by: Gal Hubara Agam <96368689+galagam@users.noreply.github.com>
3 participants
Description
nvfp4 + flashinfer_ssm mamba backend are incompatible.
nanov3 and superv3 example configs were recently updated to use flashinfer_ssm, breaking the nvfp4 support.
This is a temporary WAR. Filed bug #11455 to track. For now, fall back to triton_ssm.
Test Coverage
In follow up PR to refactor the accuracy tests to load the example config instead of a hardcoded config #11458
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