Skip to content

[None][perf] Skip KV cache estimation by default and update disagg pe…#12971

Draft
HuiGao-NV wants to merge 3 commits into
NVIDIA:mainfrom
HuiGao-NV:update_kv_cache_fraction
Draft

[None][perf] Skip KV cache estimation by default and update disagg pe…#12971
HuiGao-NV wants to merge 3 commits into
NVIDIA:mainfrom
HuiGao-NV:update_kv_cache_fraction

Conversation

@HuiGao-NV
Copy link
Copy Markdown
Collaborator

@HuiGao-NV HuiGao-NV commented Apr 12, 2026 •
edited by coderabbitai Bot
Loading

…rf-sanity fractions

Set TRTLLM_SKIP_KV_CACHE_ESTIMATION default to 1 to skip runtime KV cache memory estimation. Update free_gpu_memory_fraction in 17 disaggregated perf-sanity configs based on profiled memory usage:
ctx_new = KV_cache / (device_total - peak + KV_cache)

Summary by CodeRabbit

  • Chores
    • Updated KV cache memory estimation behavior to enable by default for improved executor initialization
    • Adjusted GPU memory allocation parameters across multiple performance benchmark configurations to optimize memory utilization

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

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

@HuiGao-NV HuiGao-NV requested a review from a team as a code owner April 12, 2026 22:34
@HuiGao-NV HuiGao-NV requested a review from lfr-0531 April 12, 2026 22:34
@HuiGao-NV HuiGao-NV marked this pull request as draft April 12, 2026 22:34
@coderabbitai
Copy link
Copy Markdown
Contributor

coderabbitai Bot commented Apr 12, 2026

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info
⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: aae5e133-d43f-400c-a6bd-7d0aa46eab9e

📥 Commits

Reviewing files that changed from the base of the PR and between 26a28ea and 4f3152b.

📒 Files selected for processing (18)
  • tensorrt_llm/_torch/pyexecutor/py_executor_creator.py
  • tests/scripts/perf-sanity/disaggregated/gb200_deepseek-r1-fp4_128k8k_con1_ctx1_pp8_gen1_tep8_eplb0_mtp3_ccb-UCX.yaml
  • tests/scripts/perf-sanity/disaggregated/gb200_deepseek-r1-fp4_1k1k_con1024_ctx1_dep4_gen1_dep8_eplb0_mtp0_ccb-UCX.yaml
  • tests/scripts/perf-sanity/disaggregated/gb200_deepseek-r1-fp4_1k1k_con1_ctx1_dep4_gen1_tep8_eplb0_mtp3_ccb-UCX.yaml
  • tests/scripts/perf-sanity/disaggregated/gb200_deepseek-r1-fp4_1k1k_con2048_ctx2_dep4_gen1_dep16_eplb0_mtp3_ccb-UCX.yaml
  • tests/scripts/perf-sanity/disaggregated/gb200_deepseek-r1-fp4_1k1k_con2048_ctx2_dep4_gen1_dep16_eplb288_mtp3_ccb-UCX.yaml
  • tests/scripts/perf-sanity/disaggregated/gb200_deepseek-r1-fp4_1k1k_con3072_ctx1_dep4_gen1_dep4_eplb0_mtp1_ccb-UCX.yaml
  • tests/scripts/perf-sanity/disaggregated/gb200_deepseek-r1-fp4_8k1k_con1_ctx1_dep4_gen1_tep8_eplb0_mtp3_ccb-UCX.yaml
  • tests/scripts/perf-sanity/disaggregated/gb200_deepseek-v32-fp4_1k1k_con2048_ctx1_dep4_gen1_dep4_eplb0_mtp1_ccb-UCX.yaml
  • tests/scripts/perf-sanity/disaggregated/gb200_gpt-oss-120b-fp4_1k1k_con2048_ctx1_tp1_gen1_dep2_eplb0_mtp0_ccb-UCX.yaml
  • tests/scripts/perf-sanity/disaggregated/gb200_gpt-oss-120b-fp4_1k1k_con512_ctx1_tp1_gen1_dep2_eplb0_mtp0_ccb-UCX.yaml
  • tests/scripts/perf-sanity/disaggregated/gb200_gpt-oss-120b-fp4_1k1k_con64_ctx1_tp1_gen1_tp4_eplb0_mtp0_ccb-UCX.yaml
  • tests/scripts/perf-sanity/disaggregated/gb200_gpt-oss-120b-fp4_8k1k_con128_ctx1_tp1_gen1_tp4_eplb0_mtp0_ccb-UCX.yaml
  • tests/scripts/perf-sanity/disaggregated/gb200_gpt-oss-120b-fp4_8k1k_con4_ctx1_tp1_gen1_tp4_eplb0_mtp0_ccb-UCX.yaml
  • tests/scripts/perf-sanity/disaggregated/gb200_gpt-oss-120b-fp4_8k1k_con512_ctx1_tp1_gen1_dep2_eplb0_mtp0_ccb-UCX.yaml
  • tests/scripts/perf-sanity/disaggregated/gb200_kimi-k2-thinking-fp4_1k1k_con4_ctx1_dep4_gen1_tep4_eplb0_mtp0_ccb-UCX.yaml
  • tests/scripts/perf-sanity/disaggregated/gb200_qwen3-235b-fp4_8k1k_con1024_ctx1_tp1_gen1_dep8_eplb0_mtp0_ccb-UCX.yaml
  • tests/scripts/perf-sanity/disaggregated/gb200_qwen3-235b-fp4_8k1k_con64_ctx1_tp1_gen1_tep4_eplb0_mtp0_ccb-UCX.yaml
📝 Walkthrough

Walkthrough

Modified KV-cache estimation default behavior by changing the environment variable fallback from '0' to '1', and updated GPU memory fraction allocations across 18 disaggregated test configuration files for various model deployments.

Changes

Cohort / File(s) Summary
KV Cache Estimation Default
tensorrt_llm/_torch/pyexecutor/py_executor_creator.py		 Changed TRTLLM_SKIP_KV_CACHE_ESTIMATION default from '0' to '1', altering the skip_est parameter passed to KvCacheCreator to enable KV cache estimation skipping by default.
Deepseek-R1 FP4 Configs
tests/scripts/perf-sanity/disaggregated/gb200_deepseek-r1-fp4_*		 Reduced free_gpu_memory_fraction values across 5 configuration variants for generator and context worker KV cache allocations (e.g., 0.8→0.62, 0.9→0.61, 0.75→0.43).
Deepseek-V32 FP4 Config
tests/scripts/perf-sanity/disaggregated/gb200_deepseek-v32-fp4_1k1k_con2048_ctx1_dep4_gen1_dep4_eplb0_mtp1_ccb-UCX.yaml		 Adjusted generator and context KV cache free_gpu_memory_fraction from 0.8→0.56 and 0.6→0.42 respectively.
GPT-OSS-120B FP4 Configs
tests/scripts/perf-sanity/disaggregated/gb200_gpt-oss-120b-fp4_*		 Lowered free_gpu_memory_fraction across 6 configuration variants, predominantly from 0.9→0.48 and 0.6→0.42 for generator and context workers.
Kimi-K2-Thinking FP4 Config
tests/scripts/perf-sanity/disaggregated/gb200_kimi-k2-thinking-fp4_4k1k_con4_ctx1_dep4_gen1_tep4_eplb0_mtp0_ccb-UCX.yaml		 Adjusted KV cache free_gpu_memory_fraction from 0.8→0.66 (gen) and 0.6→0.49 (ctx).
Qwen3-235B FP4 Configs
tests/scripts/perf-sanity/disaggregated/gb200_qwen3-235b-fp4_*		 Updated free_gpu_memory_fraction across 2 configuration variants from 0.9→0.59 (gen) and 0.6→0.39 (ctx).

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~25 minutes

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 inconclusive)

Check name Status Explanation Resolution
Description check ❓ Inconclusive The PR description provides the basic context (TRTLLM_SKIP_KV_CACHE_ESTIMATION change and config updates) but leaves critical template sections incomplete or unfilled. Complete the 'Description' and 'Test Coverage' sections to explain the rationale and testing approach; verify PR checklist items are addressed appropriately.
✅ Passed checks (2 passed)
Check name Status Explanation
Title check ✅ Passed The title clearly summarizes the main changes: skipping KV cache estimation by default and updating disaggregated perf-sanity configuration fractions.
Docstring Coverage ✅ Passed Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.

✏️ Tip: You can configure your own custom pre-merge checks in the settings.

✨ Finishing Touches
🧪 Generate unit tests (beta)
  • Create PR with unit tests

Comment @coderabbitai help to get the list of available commands and usage tips.

@HuiGao-NV
Copy link
Copy Markdown
Collaborator Author

HuiGao-NV commented Apr 12, 2026

/bot run --stage-list="GB200-12_GPUs-3_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU1-GEN1-NODE2-GPU8,GB200-12_GPUs-3_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE2-GPU8,GB200-16_GPUs-4_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE2-GPU8-GEN1-NODE2-GPU8,GB200-24_GPUs-6_Nodes-PyTorch-Disagg-PerfSanity-CTX2-NODE1-GPU4-GEN1-NODE4-GPU16,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU1-GEN1-NODE1-GPU2,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU1-GEN1-NODE1-GPU4,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4" --disable-fail-fast

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 12, 2026

PR_Github #42901 [ run ] triggered by Bot. Commit: 4f3152b Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 12, 2026

PR_Github #42901 [ run ] completed with state FAILURE. Commit: 4f3152b
/LLM/main/L0_MergeRequest_PR pipeline #33563 (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

Link to invocation

@HuiGao-NV
Copy link
Copy Markdown
Collaborator Author

HuiGao-NV commented Apr 13, 2026

/bot run --stage-list="GB200-12_GPUs-3_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU1-GEN1-NODE2-GPU8,GB200-12_GPUs-3_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE2-GPU8,GB200-16_GPUs-4_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE2-GPU8-GEN1-NODE2-GPU8,GB200-24_GPUs-6_Nodes-PyTorch-Disagg-PerfSanity-CTX2-NODE1-GPU4-GEN1-NODE4-GPU16,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU1-GEN1-NODE1-GPU2,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU1-GEN1-NODE1-GPU4,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4" --disable-fail-fast

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 13, 2026

PR_Github #42924 [ run ] triggered by Bot. Commit: e58cbbf Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 13, 2026

PR_Github #42924 [ run ] completed with state FAILURE. Commit: e58cbbf
/LLM/main/L0_MergeRequest_PR pipeline #33585 (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

Link to invocation

@HuiGao-NV
Copy link
Copy Markdown
Collaborator Author

HuiGao-NV commented Apr 13, 2026

/bot run --stage-list "GB200-12_GPUs-3_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU1-GEN1-NODE2-GPU8-Post-Merge-1,GB200-12_GPUs-3_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE2-GPU8-Post-Merge-1,GB200-24_GPUs-6_Nodes-PyTorch-Disagg-PerfSanity-CTX2-NODE1-GPU4-GEN1-NODE4-GPU16-Post-Merge-4,GB200-24_GPUs-6_Nodes-PyTorch-Disagg-PerfSanity-CTX2-NODE1-GPU4-GEN1-NODE4-GPU16-Post-Merge-2,GB200-24_GPUs-6_Nodes-PyTorch-Disagg-PerfSanity-CTX2-NODE1-GPU4-GEN1-NODE4-GPU16-Post-Merge-3," --disable-fail-fast

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 13, 2026

PR_Github #42938 [ run ] triggered by Bot. Commit: 71fefce Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 13, 2026

PR_Github #42938 [ run ] completed with state SUCCESS. Commit: 71fefce
/LLM/main/L0_MergeRequest_PR pipeline #33598 (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

Link to invocation

@HuiGao-NV
Copy link
Copy Markdown
Collaborator Author

HuiGao-NV commented Apr 14, 2026

/bot run --stage-list "GB200-12_GPUs-3_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU1-GEN1-NODE2-GPU8-Post-Merge-1,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-2" --disable-fail-fast

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 14, 2026

PR_Github #43192 [ run ] triggered by Bot. Commit: 56d1092 Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 14, 2026

PR_Github #43192 [ run ] completed with state DISABLED
Freeze main and open the PR merge only after CI is back to healthy https://nvidia.slack.com/archives/C059LSY62BT/p1776141760843319?thread_ts=1775985925.442509&cid=C059LSY62BT

Link to invocation

@HuiGao-NV HuiGao-NV force-pushed the update_kv_cache_fraction branch from 56d1092 to 2078af8 Compare April 16, 2026 11:00
@github-actions
Copy link
Copy Markdown

github-actions Bot commented Apr 16, 2026

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. 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 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.

@HuiGao-NV HuiGao-NV force-pushed the update_kv_cache_fraction branch from 30c7e37 to 3dafac1 Compare April 16, 2026 11:08
@HuiGao-NV
Copy link
Copy Markdown
Collaborator Author

HuiGao-NV commented Apr 16, 2026

/bot run --post-merge --only-multi-gpu-test --disable-fail-fast

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 16, 2026

PR_Github #43766 [ run ] triggered by Bot. Commit: e449997 Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 16, 2026

PR_Github #43766 [ run ] completed with state FAILURE. Commit: e449997
/LLM/main/L0_MergeRequest_PR pipeline #34247 (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

Link to invocation

@HuiGao-NV
Copy link
Copy Markdown
Collaborator Author

HuiGao-NV commented Apr 17, 2026

/bot run --stage-list "GB200-12_GPUs-3_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU1-GEN1-NODE2-GPU8-Post-Merge-1,GB200-12_GPUs-3_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE2-GPU8-Post-Merge-1,GB200-12_GPUs-3_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE2-GPU8-Post-Merge-3,GB200-12_GPUs-3_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE2-GPU8-Post-Merge-6,GB200-20_GPUs-5_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE4-GPU16-Post-Merge-1,GB200-36_GPUs-9_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE8-GPU32-Post-Merge-3,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU1-GEN1-NODE1-GPU2-Post-Merge-1,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU1-GEN1-NODE1-GPU2-Post-Merge-3,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-1,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-2,GB200-8_GPUs-2_Nodes-PyTorch-PerfSanity-Node2-GPU8-Post-Merge-6" --disable-fail-fast

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 17, 2026

PR_Github #43901 [ run ] triggered by Bot. Commit: e053b7a Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 17, 2026

PR_Github #43901 [ run ] completed with state FAILURE. Commit: e053b7a
/LLM/main/L0_MergeRequest_PR pipeline #34351 (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

Link to invocation

@HuiGao-NV
Copy link
Copy Markdown
Collaborator Author

HuiGao-NV commented Apr 17, 2026

/bot run --stage-list "GB200-12_GPUs-3_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU1-GEN1-NODE2-GPU8-Post-Merge-1,GB200-12_GPUs-3_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE2-GPU8-Post-Merge-1,GB200-12_GPUs-3_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE2-GPU8-Post-Merge-3,GB200-12_GPUs-3_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE2-GPU8-Post-Merge-6,GB200-20_GPUs-5_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE4-GPU16-Post-Merge-1,GB200-36_GPUs-9_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE8-GPU32-Post-Merge-3,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU1-GEN1-NODE1-GPU2-Post-Merge-1,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU1-GEN1-NODE1-GPU2-Post-Merge-3,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-1,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-2,GB200-8_GPUs-2_Nodes-PyTorch-PerfSanity-Node2-GPU8-Post-Merge-6" --disable-fail-fast

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 17, 2026

PR_Github #43975 [ run ] triggered by Bot. Commit: e053b7a Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 17, 2026

PR_Github #43975 [ run ] completed with state FAILURE. Commit: e053b7a
/LLM/main/L0_MergeRequest_PR pipeline #34416 (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

Link to invocation

@HuiGao-NV HuiGao-NV force-pushed the update_kv_cache_fraction branch from e053b7a to 0f6b9c8 Compare April 19, 2026 03:44
@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 30, 2026

PR_Github #46316 [ run ] triggered by Bot. Commit: 50c3772 Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 30, 2026

PR_Github #46316 [ run ] completed with state SUCCESS. Commit: 50c3772
/LLM/main/L0_MergeRequest_PR pipeline #36415 (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

Link to invocation

@HuiGao-NV
Copy link
Copy Markdown
Collaborator Author

HuiGao-NV commented Apr 30, 2026

/bot run --post-merge --disable-multi-gpu-test --disable-fail-fast

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 30, 2026

PR_Github #46359 [ run ] triggered by Bot. Commit: 50c3772 Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented May 1, 2026

PR_Github #46359 [ run ] completed with state ABORTED. Commit: 50c3772

Link to invocation

@HuiGao-NV
Copy link
Copy Markdown
Collaborator Author

HuiGao-NV commented May 2, 2026

/bot run --stage-list="DGX_B200-4_GPUs-PyTorch-Post-Merge-2,DGX_B200-4_GPUs-PyTorch-Post-Merge-1,GB200-20_GPUs-5_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE4-GPU16-Post-Merge-4,GB200-36_GPUs-9_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE8-GPU32-Post-Merge-8,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-1,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-2,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-3,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-4,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-7,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-3,GB200-8_GPUs-2_Nodes-PyTorch-PerfSanity-Node2-GPU8-Post-Merge-3,GB300-4_GPUs-PyTorch-PerfSanity-Post-Merge-1" --disable-fail-fast

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented May 2, 2026

PR_Github #46532 [ run ] triggered by Bot. Commit: 50c3772 Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented May 2, 2026

PR_Github #46532 [ run ] completed with state FAILURE. Commit: 50c3772
/LLM/main/L0_MergeRequest_PR pipeline #36591 (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

Link to invocation

@HuiGao-NV HuiGao-NV force-pushed the update_kv_cache_fraction branch from 50c3772 to d566451 Compare May 3, 2026 01:33
@HuiGao-NV
Copy link
Copy Markdown
Collaborator Author

HuiGao-NV commented May 3, 2026

/bot run --stage-list="DGX_B200-4_GPUs-PyTorch-Post-Merge-2,DGX_B200-4_GPUs-PyTorch-Post-Merge-1,GB200-20_GPUs-5_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE4-GPU16-Post-Merge-4,GB200-36_GPUs-9_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE8-GPU32-Post-Merge-8,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-1,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-2,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-3,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-4,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-7,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-3,GB200-8_GPUs-2_Nodes-PyTorch-PerfSanity-Node2-GPU8-Post-Merge-3,GB300-4_GPUs-PyTorch-PerfSanity-Post-Merge-1" --disable-fail-fast

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented May 3, 2026

PR_Github #46572 [ run ] triggered by Bot. Commit: d566451 Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented May 3, 2026

PR_Github #46572 [ run ] completed with state FAILURE. Commit: d566451
/LLM/main/L0_MergeRequest_PR pipeline #36623 (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

@HuiGao-NV
Copy link
Copy Markdown
Collaborator Author

HuiGao-NV commented May 3, 2026

/bot run --stage-list="DGX_B200-4_GPUs-PyTorch-Post-Merge-2,DGX_B200-4_GPUs-PyTorch-Post-Merge-1,GB200-20_GPUs-5_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE4-GPU16-Post-Merge-4,GB200-36_GPUs-9_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE8-GPU32-Post-Merge-8,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-1,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-2,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-4,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-7,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-3,GB200-8_GPUs-2_Nodes-PyTorch-PerfSanity-Node2-GPU8-Post-Merge-3,GB300-4_GPUs-PyTorch-PerfSanity-Post-Merge-1,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-1,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-2,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-4,GB200-8_GPUs-2_Nodes-PyTorch-PerfSanity-Node2-GPU8-Post-Merge-6" --disable-fail-fast

@github-actions
Copy link
Copy Markdown

github-actions Bot commented May 3, 2026

⚠️ Bot command ignored: The /bot command must appear at the very beginning of the comment (no leading blank lines or spaces). Please post a new comment with /bot as the first character.

@HuiGao-NV
Copy link
Copy Markdown
Collaborator Author

HuiGao-NV commented May 4, 2026

/bot run --stage-list="DGX_B200-4_GPUs-PyTorch-Post-Merge-2,DGX_B200-4_GPUs-PyTorch-Post-Merge-1,GB200-20_GPUs-5_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE4-GPU16-Post-Merge-4,GB200-36_GPUs-9_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE8-GPU32-Post-Merge-8,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-1,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-2,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-4,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-7,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-3,GB200-8_GPUs-2_Nodes-PyTorch-PerfSanity-Node2-GPU8-Post-Merge-3,GB300-4_GPUs-PyTorch-PerfSanity-Post-Merge-1,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-1,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-2,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-4,GB200-8_GPUs-2_Nodes-PyTorch-PerfSanity-Node2-GPU8-Post-Merge-6" --disable-fail-fast

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented May 4, 2026

PR_Github #46614 [ run ] triggered by Bot. Commit: d3995a8 Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented May 4, 2026

PR_Github #46614 [ run ] completed with state FAILURE. Commit: d3995a8
/LLM/main/L0_MergeRequest_PR pipeline #36661 (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

@HuiGao-NV
Update kv cache fraction
e76d2cc 
Signed-off-by: Hui Gao <huig@nvidia.com>
@HuiGao-NV HuiGao-NV force-pushed the update_kv_cache_fraction branch from d3995a8 to e76d2cc Compare May 5, 2026 00:38
@HuiGao-NV
Copy link
Copy Markdown
Collaborator Author

HuiGao-NV commented May 5, 2026

/bot run --stage-list="DGX_B200-4_GPUs-PyTorch-Post-Merge-2,GB200-36_GPUs-9_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE8-GPU32-Post-Merge-8,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-1,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-2,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-7,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-3,GB200-8_GPUs-2_Nodes-PyTorch-PerfSanity-Node2-GPU8-Post-Merge-3,GB300-4_GPUs-PyTorch-PerfSanity-Post-Merge-1,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-1,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-2,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-4" --disable-fail-fast

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented May 5, 2026

PR_Github #46709 [ run ] triggered by Bot. Commit: e76d2cc Link to invocation

@HuiGao-NV
Fix oom errors in CI
a481aa7 
Signed-off-by: Hui Gao <huig@nvidia.com>
@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented May 5, 2026

PR_Github #46709 [ run ] completed with state FAILURE. Commit: e76d2cc
/LLM/main/L0_MergeRequest_PR pipeline #36744 (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

@HuiGao-NV
Copy link
Copy Markdown
Collaborator Author

HuiGao-NV commented May 5, 2026

/bot run --stage-list="GB200-36_GPUs-9_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE8-GPU32-Post-Merge-8,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-2,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-7,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-3,GB200-8_GPUs-2_Nodes-PyTorch-PerfSanity-Node2-GPU8-Post-Merge-3,GB300-4_GPUs-PyTorch-PerfSanity-Post-Merge-1,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-1,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-2,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-4" --disable-fail-fast

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented May 5, 2026

PR_Github #46749 [ run ] triggered by Bot. Commit: a481aa7 Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented May 5, 2026

PR_Github #46749 [ run ] completed with state FAILURE. Commit: a481aa7
/LLM/main/L0_MergeRequest_PR pipeline #36777 (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

@HuiGao-NV
Update kv cache fraction for new cases
674ff55 
Signed-off-by: Hui Gao <huig@nvidia.com>
@HuiGao-NV
Copy link
Copy Markdown
Collaborator Author

HuiGao-NV commented May 6, 2026

/bot run --stage-list="GB200-36_GPUs-9_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE8-GPU32-Post-Merge-8,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-2,GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-7,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-3,GB200-8_GPUs-2_Nodes-PyTorch-PerfSanity-Node2-GPU8-Post-Merge-3,GB300-4_GPUs-PyTorch-PerfSanity-Post-Merge-1,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-1,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-2,GB200-8_GPUs-2_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE1-GPU4-Post-Merge-4" --disable-fail-fast

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented May 6, 2026

PR_Github #46878 [ run ] triggered by Bot. Commit: 674ff55 Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented May 6, 2026

PR_Github #46878 [ run ] completed with state FAILURE. Commit: 674ff55
/LLM/main/L0_MergeRequest_PR pipeline #36889 (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

Link to invocation

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@lfr-0531 lfr-0531 Awaiting requested review from lfr-0531 lfr-0531 is a code owner automatically assigned from NVIDIA/trt-llm-torch-runtime-devs

At least 1 approving review is required to merge this pull request.

Assignees

@HuiGao-NV HuiGao-NV

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@HuiGao-NV @tensorrt-cicd