TensorRT-LLM
TensorRT-LLM copied to clipboard
NVIDIA
[nvbug/5285881][fix] Fix chunked prefill + overlap scheduler
[nvbug/5285881][fix] Fix chunked prefill + overlap scheduler
Description
The token_idx needed to be advanced when the decoder processes chunked requests during overlap scheduling
Test Coverage
Unwaived existing tests. Overlap scheduling getting turned on by default broke these tests. Confirmed they pass locally.
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 [--disable-fail-fast --skip-test --stage-list "A10-1, xxx" --gpu-type "A30, H100_PCIe" --add-multi-gpu-test --only-multi-gpu-test --disable-multi-gpu-test --post-merge --extra-stage "H100_PCIe-[Post-Merge]-1, xxx"]
Launch build/test pipelines. All previously running jobs will be killed.
--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 "A10-1, xxx" (OPTIONAL) : Only run the specified test stages. Examples: "A10-1, 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.
--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. Will also run 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-[Post-Merge]-1, xxx" (OPTIONAL) : Run the ordinary L0 pre-merge pipeline and specified test stages. Examples: --extra-stage "H100_PCIe-[Post-Merge]-1, xxx".
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.
PR_Github #5530 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #4031 completed with status: 'FAILURE'
PR_Github #5738 [ run ] completed with state FAILURE
/LLM/main/L0_MergeRequest_PR pipeline #4194 completed with status: 'FAILURE'
PR_Github #5773 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #4227 completed with status: 'FAILURE'
PR_Github #5877 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #4307 completed with status: 'FAILURE'
The fix was already merged into main. Someone evidently ran into this bug and fixed it before the nvbug ever ended up on my desk; it just took a week or two to get merged due to CI. So now, this PR only contains tests
PR_Github #6034 [ run ] completed with state FAILURE
/LLM/main/L0_MergeRequest_PR pipeline #4409 completed with status: 'FAILURE'
PR_Github #6055 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #4427 completed with status: 'FAILURE'
PR_Github #6163 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #4506 completed with status: 'FAILURE'
/bot skip --comment "Confirmed that both of the newly added chunked prefill tests pass. There are no other code changes in this PR"