[None][test] Split test_moe_backend TRTLLM by quant=None to avoid in-process IMA cascade by leslie-fang25 · Pull Request #16068 · NVIDIA/TensorRT-LLM

leslie-fang25 · 2026-07-07T12:38:04Z

Split test_moe_backend TRTLLM by quant=None to avoid in-process IMA cascade

Summary by CodeRabbit

Tests
- Updated test selection for one backend test to split TRTLLM coverage into separate variants, improving how different conditions are exercised.
- Adjusted waiver entries for related end-to-end tests to improve organization and readability.

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)
If PR introduces API changes, an appropriate PR label is added - either api-compatible or api-breaking. For api-breaking, include BREAKING in the PR title.
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

To see a list of available CI bot commands, please comment /bot help.

coderabbitai · 2026-07-07T12:41:09Z

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: fbba2b6a-0a38-4fef-b170-92d3a45516cb

📥 Commits

Reviewing files that changed from the base of the PR and between ce67288 and e80e813.

📒 Files selected for processing (3)

tests/integration/test_lists/test-db/l0_b200.yml
tests/integration/test_lists/test-db/l0_b300.yml
tests/integration/test_lists/waives.txt

💤 Files with no reviewable changes (1)

tests/integration/test_lists/waives.txt

📝 Walkthrough

Walkthrough

This PR updates test list configuration files for B200 and B300 GPU test databases, splitting a single TRTLLM filter for the test_moe_backend test into two separate filtered entries ("TRTLLM and not None" and "TRTLLM and None"). It also reformats and removes a corresponding waiver entry in waives.txt.

Changes

Test Selection Updates

Layer / File(s)	Summary
Split TRTLLM filter in test-db lists `tests/integration/test_lists/test-db/l0_b200.yml`, `tests/integration/test_lists/test-db/l0_b300.yml`	Replaces the single `-k "TRTLLM"` filter for `test_moe_backend` with two entries filtered by `TRTLLM and not None` and `TRTLLM and None`.
Update waives.txt entries `tests/integration/test_lists/waives.txt`	Adds blank-line separators between three DGX_B200 `test_skip_softmax_attention_4gpus` waivers and removes the neighboring B300 `test_moe_backend` TRTLLM waiver.

Estimated code review effort: 1 (Trivial) | ~5 minutes

Possibly related PRs

NVIDIA/TensorRT-LLM#15959: Both PRs modify tests/integration/test_lists/waives.txt and affect which test_moe_backend.py TRTLLM-related variants are skipped.

Suggested reviewers: yizhang-nv, yechank-nvidia, PerkzZheng

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The description keeps the template headings but leaves Description and Test Coverage empty, so the rationale and validation are missing.	Add a short Description explaining the issue and fix, plus Test Coverage listing the relevant tests or CI evidence.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title is concise and matches the main change: splitting test_moe_backend TRTLLM by quant=None to avoid the IMA cascade.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests

⚔️ Resolve merge conflicts

Resolve merge conflict in branch leslie/unwaive_some_trtllm_moe_test_case

_{Comment @coderabbitai help to get the list of available commands.}

leslie-fang25 · 2026-07-07T12:44:38Z

/bot run --disable-fail-fast

tensorrt-cicd · 2026-07-07T12:51:30Z

PR_Github #58003 [ run ] triggered by Bot. Commit: 5e40e0d Link to invocation

tensorrt-cicd · 2026-07-07T13:29:12Z

PR_Github #58003 [ run ] completed with state SUCCESS. Commit: 5e40e0d
/LLM/main/L0_MergeRequest_PR pipeline #46673 completed with status: 'SUCCESS'

CI Report

Link to invocation

leslie-fang25 · 2026-07-07T23:50:14Z

/bot run --disable-fail-fast --extra-stage "B300-PyTorch-Post-Merge-1,DGX_B200-PyTorch-Post-Merge-1"

tensorrt-cicd · 2026-07-07T23:56:49Z

PR_Github #58095 [ run ] triggered by Bot. Commit: 5e40e0d Link to invocation

tensorrt-cicd · 2026-07-08T04:19:24Z

PR_Github #58095 [ run ] completed with state SUCCESS. Commit: 5e40e0d
/LLM/main/L0_MergeRequest_PR pipeline #46758 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

leslie-fang25 · 2026-07-08T05:22:43Z

/bot help

github-actions · 2026-07-08T05:22:51Z

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) --high-priority]

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. Supports wildcard * for pattern matching (e.g., "*PerfSanity*" matches all stages containing PerfSanity). Examples: "A10-PyTorch-1, xxx", "PerfSanity". 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. Supports wildcard * for pattern matching. Examples: --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx", --extra-stage "Post-Merge".

--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 the stage-list parameter to access the appropriate container environment. Note: Does NOT update GitHub check status.

--high-priority (OPTIONAL) : Run the pipeline with high priority. This option is restricted to authorized users only and will route the job to a high-priority queue.

kill

kill

Kill all running builds associated with pull request.

skip

skip --comment COMMENT

Skip 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-pipeline

Reuse 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.

leslie-fang25 · 2026-07-08T05:23:32Z

/bot run --disable-reuse-test --disable-fail-fast --post-merge --stage-list “B300-PyTorch-Post-Merge-1,DGX_B200-PyTorch-Post-Merge-1,DGX_B200-PyTorch-Post-Merge-2”

leslie-fang25 · 2026-07-08T05:39:24Z

/bot run --disable-reuse-test --disable-fail-fast --post-merge --stage-list “B300-PyTorch-Post-Merge-1,DGX_B200-PyTorch-Post-Merge-1,DGX_B200-PyTorch-Post-Merge-2”

tensorrt-cicd · 2026-07-08T05:46:12Z

PR_Github #58159 [ run ] triggered by Bot. Commit: 5e40e0d Link to invocation

tensorrt-cicd · 2026-07-08T06:22:26Z

PR_Github #58159 [ run ] completed with state FAILURE. Commit: 5e40e0d
/LLM/main/L0_MergeRequest_PR pipeline #46810 (Partly Tested) completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

leslie-fang25 · 2026-07-08T06:31:28Z

/bot run --disable-reuse-test --disable-fail-fast --post-merge --stage-list “B300-PyTorch-Post-Merge-1, DGX_B200-PyTorch-Post-Merge-1, DGX_B200-PyTorch-Post-Merge-2”

tensorrt-cicd · 2026-07-08T06:37:05Z

PR_Github #58180 Bot args parsing error: usage: /bot [-h]
{run,kill,skip,submit,reviewers,reuse-pipeline,reuse-review} ...
/bot: error: unrecognized arguments: DGX_B200-PyTorch-Post-Merge-1, DGX_B200-PyTorch-Post-Merge-2”

Link to invocation

leslie-fang25 · 2026-07-08T06:47:34Z

/bot run --disable-reuse-test --disable-fail-fast --post-merge --stage-list "B300-PyTorch-Post-Merge-1,DGX_B200-PyTorch-Post-Merge-1,DGX_B200-PyTorch-Post-Merge-2"

tensorrt-cicd · 2026-07-08T06:52:59Z

PR_Github #58184 [ run ] triggered by Bot. Commit: 5e40e0d Link to invocation

tensorrt-cicd · 2026-07-08T11:01:35Z

PR_Github #58184 [ run ] completed with state FAILURE. Commit: 5e40e0d
/LLM/main/L0_MergeRequest_PR pipeline #46832 (Partly Tested) completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

leslie-fang25 · 2026-07-08T12:02:01Z

/bot run --disable-fail-fast --post-merge --stage-list "B300-PyTorch-Post-Merge-1,DGX_B200-PyTorch-Post-Merge-1,DGX_B200-PyTorch-Post-Merge-2"

tensorrt-cicd · 2026-07-08T12:07:31Z

PR_Github #58229 [ run ] triggered by Bot. Commit: 5e40e0d Link to invocation

tensorrt-cicd · 2026-07-08T14:00:25Z

PR_Github #58229 [ run ] completed with state FAILURE. Commit: 5e40e0d
/LLM/main/L0_MergeRequest_PR pipeline #46871 (Partly Tested) completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

leslie-fang25 · 2026-07-08T23:45:28Z

/bot run --disable-fail-fast --post-merge --stage-list "B300-PyTorch-Post-Merge-1,DGX_B200-PyTorch-Post-Merge-1,DGX_B200-PyTorch-Post-Merge-2"

tensorrt-cicd · 2026-07-08T23:52:35Z

PR_Github #58325 [ run ] triggered by Bot. Commit: 5e40e0d Link to invocation

tensorrt-cicd · 2026-07-09T01:34:07Z

PR_Github #58325 [ run ] completed with state SUCCESS. Commit: 5e40e0d
/LLM/main/L0_MergeRequest_PR pipeline #46955 (Partly Tested) completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

leslie-fang25 · 2026-07-09T02:15:18Z

/bot run --disable-fail-fast --post-merge --stage-list "B300-PyTorch-Post-Merge-1,DGX_B200-PyTorch-Post-Merge-1,DGX_B200-PyTorch-Post-Merge-2"

tensorrt-cicd · 2026-07-09T02:20:55Z

PR_Github #58347 [ run ] triggered by Bot. Commit: 5e40e0d Link to invocation

tensorrt-cicd · 2026-07-09T04:11:46Z

PR_Github #58347 [ run ] completed with state SUCCESS. Commit: 5e40e0d
/LLM/main/L0_MergeRequest_PR pipeline #46976 (Partly Tested) completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

leslie-fang25 · 2026-07-09T08:40:50Z

/bot run --disable-reuse-test --disable-fail-fast --post-merge --stage-list "B300-PyTorch-Post-Merge-1,DGX_B200-PyTorch-Post-Merge-*"

leslie-fang25 · 2026-07-13T10:37:20Z

/bot run --disable-reuse-test --disable-fail-fast --post-merge --stage-list "B300-PyTorch-Post-Merge-1,DGX_B200-PyTorch-Post-Merge-*"

…RTLLM MoE cases on Blackwell Signed-off-by: leslie-fang25 <leslief@nvidia.com>

…n Blackwell The blanket B300/DGX_B200 waives are superseded by the in-process cascade skips added in should_skip_trtllm (quant=None and e128 W4A16_MXFP4); the remaining TRTLLM MoE cases now run in CI. Signed-off-by: leslie-fang25 <leslief@nvidia.com>

leslie-fang25 · 2026-07-13T11:09:42Z

/bot run --disable-reuse-test --disable-fail-fast --post-merge --stage-list "B300-PyTorch-Post-Merge-1,DGX_B200-PyTorch-Post-Merge-*"

tensorrt-cicd · 2026-07-13T11:15:37Z

PR_Github #58964 [ run ] triggered by Bot. Commit: a554b2f Link to invocation

tensorrt-cicd · 2026-07-13T12:57:06Z

PR_Github #58964 [ run ] completed with state SUCCESS. Commit: a554b2f
/LLM/main/L0_MergeRequest_PR pipeline #47496 (Partly Tested) completed with status: 'SUCCESS'

CI Report

Link to invocation

leslie-fang25 · 2026-07-13T12:57:28Z

/bot run --disable-reuse-test --disable-fail-fast --post-merge --stage-list "B300-PyTorch-Post-Merge-1,DGX_B200-PyTorch-Post-Merge-*"

tensorrt-cicd · 2026-07-13T13:03:00Z

PR_Github #58975 [ run ] triggered by Bot. Commit: a554b2f Link to invocation

tensorrt-cicd · 2026-07-13T15:04:37Z

PR_Github #58975 [ run ] completed with state SUCCESS. Commit: a554b2f
/LLM/main/L0_MergeRequest_PR pipeline #47506 (Partly Tested) completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

leslie-fang25 · 2026-07-13T23:25:39Z

/bot run --disable-fail-fast --post-merge --stage-list "B300-PyTorch-Post-Merge-1,DGX_B200-PyTorch-Post-Merge-*"

tensorrt-cicd · 2026-07-13T23:31:37Z

PR_Github #59044 [ run ] triggered by Bot. Commit: a554b2f Link to invocation

tensorrt-cicd · 2026-07-13T23:57:18Z

PR_Github #59044 [ run ] completed with state SUCCESS. Commit: a554b2f
/LLM/main/L0_MergeRequest_PR pipeline #47569 (Partly Tested) completed with status: 'SUCCESS'

CI Report

Link to invocation

leslie-fang25 · 2026-07-14T02:10:11Z

/bot run --disable-fail-fast

leslie-fang25 · 2026-07-14T02:11:44Z

The implementation has been modified based on offline discussion. @xxi-nv please kindly help to take a look again.

tensorrt-cicd · 2026-07-14T02:18:10Z

PR_Github #59068 [ run ] triggered by Bot. Commit: 4b644c9 Link to invocation

leslie-fang25 · 2026-07-14T05:09:28Z

/bot run --disable-fail-fast

tensorrt-cicd · 2026-07-14T05:14:47Z

PR_Github #59112 [ run ] triggered by Bot. Commit: 4b644c9 Link to invocation

tensorrt-cicd · 2026-07-14T05:19:23Z

PR_Github #59068 [ run ] completed with state ABORTED. Commit: 4b644c9

Link to invocation

tensorrt-cicd · 2026-07-14T07:31:36Z

PR_Github #59112 [ run ] completed with state SUCCESS. Commit: 4b644c9
/LLM/main/L0_MergeRequest_PR pipeline #47624 completed with status: 'SUCCESS'

CI Report

Link to invocation

github-actions Bot assigned leslie-fang25 Jul 7, 2026

leslie-fang25 force-pushed the leslie/unwaive_some_trtllm_moe_test_case branch from e80e813 to 5e40e0d Compare July 7, 2026 12:43

leslie-fang25 added 2 commits July 13, 2026 11:02

[https://nvbugs/6165866][test] Skip quant=None and e128 W4A16_MXFP4 T…

1f17260

…RTLLM MoE cases on Blackwell Signed-off-by: leslie-fang25 <leslief@nvidia.com>

leslie-fang25 force-pushed the leslie/unwaive_some_trtllm_moe_test_case branch from 917165b to a554b2f Compare July 13, 2026 11:05

leslie-fang25 force-pushed the leslie/unwaive_some_trtllm_moe_test_case branch from a554b2f to 4b644c9 Compare July 14, 2026 02:07

xxi-nv approved these changes Jul 14, 2026

View reviewed changes

leslie-fang25 requested review from a team, BowenFu, brnguyen2, fredricz-20070104 and ruodil July 15, 2026 07:09

ruodil approved these changes Jul 15, 2026

View reviewed changes

leslie-fang25 merged commit 95f9f92 into NVIDIA:main Jul 15, 2026
7 checks passed

Uh oh!

Conversation

leslie-fang25 commented Jul 7, 2026 • edited by coderabbitai Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary by CodeRabbit

Description

Test Coverage

PR Checklist

GitHub Bot Help

Uh oh!

coderabbitai Bot commented Jul 7, 2026

Walkthrough

Changes

❌ Failed checks (1 warning)

Uh oh!

leslie-fang25 commented Jul 7, 2026

Uh oh!

tensorrt-cicd commented Jul 7, 2026

Uh oh!

tensorrt-cicd commented Jul 7, 2026

Uh oh!

leslie-fang25 commented Jul 7, 2026

Uh oh!

tensorrt-cicd commented Jul 7, 2026

Uh oh!

tensorrt-cicd commented Jul 8, 2026

Uh oh!

leslie-fang25 commented Jul 8, 2026

Uh oh!

github-actions Bot commented Jul 8, 2026

GitHub Bot Help

kill

skip

reuse-pipeline

Uh oh!

leslie-fang25 commented Jul 8, 2026

Uh oh!

leslie-fang25 commented Jul 8, 2026

Uh oh!

tensorrt-cicd commented Jul 8, 2026

Uh oh!

tensorrt-cicd commented Jul 8, 2026

Uh oh!

leslie-fang25 commented Jul 8, 2026

Uh oh!

tensorrt-cicd commented Jul 8, 2026

Uh oh!

leslie-fang25 commented Jul 8, 2026

Uh oh!

tensorrt-cicd commented Jul 8, 2026

Uh oh!

tensorrt-cicd commented Jul 8, 2026

Uh oh!

leslie-fang25 commented Jul 8, 2026

Uh oh!

tensorrt-cicd commented Jul 8, 2026

Uh oh!

tensorrt-cicd commented Jul 8, 2026

Uh oh!

leslie-fang25 commented Jul 8, 2026

Uh oh!

tensorrt-cicd commented Jul 8, 2026

Uh oh!

tensorrt-cicd commented Jul 9, 2026

Uh oh!

leslie-fang25 commented Jul 9, 2026

Uh oh!

tensorrt-cicd commented Jul 9, 2026

Uh oh!

tensorrt-cicd commented Jul 9, 2026

Uh oh!

leslie-fang25 commented Jul 9, 2026

Uh oh!

leslie-fang25 commented Jul 13, 2026

Uh oh!

leslie-fang25 commented Jul 13, 2026

Uh oh!

tensorrt-cicd commented Jul 13, 2026

Uh oh!

tensorrt-cicd commented Jul 13, 2026

leslie-fang25 commented Jul 7, 2026 •

edited by coderabbitai Bot

Loading