Skip to content

[TRTLLM-5530][BREAKING CHANGE] refactor: unify KvCacheConfig in LLM class for pytorch backend #5752

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: main
Choose a base branch
from

Conversation

Superjomn
Copy link
Collaborator

PR title

Please write the PR title by following template:

[JIRA ticket link/nvbug link/github issue link][fix/feat/doc/infra/...] <summary of this PR>

For example, assume I have a PR hope to support a new feature about cache manager of Jira TRTLLM-1000 ticket, it would be like

[TRTLLM-1000][feat] Support a new feature about cache manager

Description

Please explain the issue and the solution in short.

Test Coverage

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".

For guidance on mapping tests to stage names, see docs/source/reference/ci-overview.md.

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.

@Superjomn Superjomn requested a review from a team as a code owner July 4, 2025 07:16
@Superjomn Superjomn changed the title [BREAKING CHANGE]: unify KvCacheConfig in LLM class for pytorch backend [TRTLLM-5530][BREAKING CHANGE] refactor: unify KvCacheConfig in LLM class for pytorch backend Jul 4, 2025
@Superjomn
Copy link
Collaborator Author

/bot run --disable-fail-fast

@tensorrt-cicd
Copy link
Collaborator

PR_Github #10985 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #10985 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #8114 completed with status: 'FAILURE'

@Superjomn Superjomn force-pushed the change-kvcache-config branch from 5df018b to d0f6cb0 Compare July 6, 2025 00:06
@Superjomn
Copy link
Collaborator Author

/bot run

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11041 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11041 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #8162 completed with status: 'FAILURE'

@Superjomn Superjomn force-pushed the change-kvcache-config branch 2 times, most recently from 6abe190 to bf0839b Compare July 6, 2025 07:05
@Superjomn
Copy link
Collaborator Author

/bot run --disable-fail-fast

1 similar comment
@Superjomn
Copy link
Collaborator Author

/bot run --disable-fail-fast

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11089 [ run ] triggered by Bot

@Superjomn
Copy link
Collaborator Author

/bot kill

@Superjomn Superjomn force-pushed the change-kvcache-config branch from bf0839b to f9c7310 Compare July 7, 2025 02:59
@Superjomn Superjomn requested review from QiJune and lucaslie July 7, 2025 02:59
@tensorrt-cicd
Copy link
Collaborator

PR_Github #11091 [ kill ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11089 [ run ] completed with state ABORTED

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11091 [ kill ] completed with state SUCCESS
Successfully killed previous jobs for commit f9c7310

@Superjomn
Copy link
Collaborator Author

/bot run --disable-fail-fast

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11112 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11112 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #8218 completed with status: 'FAILURE'

@Superjomn Superjomn force-pushed the change-kvcache-config branch from f9c7310 to 67a8191 Compare July 7, 2025 14:12
@Superjomn
Copy link
Collaborator Author

/bot run --disable-fail-fast

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11158 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11158 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #8251 completed with status: 'FAILURE'

@@ -89,6 +89,9 @@ def get_settings(params: dict, dataset_metadata: DatasetMetadata, model: str,
if extra_llm_api_options:
with open(extra_llm_api_options, 'r') as f:
llm_args_dict = yaml.safe_load(f)
if "kv_cache_config" in llm_args_dict:
kv_cache_dtype = llm_args_dict["kv_cache_config"].get(
"dtype", "auto")

if "kv_cache_dtype" in llm_args_dict:
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Shouldn't we just augment the retrieval of the dtype in this conditional?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yeah, the "kv_cache_dtype" is merged to KvCacheConfig (a pure python config) as "dtype", so here it tries to fetch it from the kv_cache_config.

@Superjomn Superjomn force-pushed the change-kvcache-config branch from 67a8191 to aff5c11 Compare July 8, 2025 07:20
@Superjomn
Copy link
Collaborator Author

/bot run

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11247 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11247 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #8318 completed with status: 'FAILURE'

@Superjomn Superjomn force-pushed the change-kvcache-config branch from aff5c11 to 57dcea6 Compare July 8, 2025 10:09
@Superjomn
Copy link
Collaborator Author

/bot run

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11281 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11281 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #8344 completed with status: 'FAILURE'

@Superjomn Superjomn force-pushed the change-kvcache-config branch from 57dcea6 to fefcca4 Compare July 8, 2025 13:42
@Superjomn
Copy link
Collaborator Author

/bot run

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11309 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11309 [ run ] completed with state FAILURE
/LLM/main/L0_MergeRequest_PR pipeline #8363 completed with status: 'FAILURE'

@Superjomn Superjomn force-pushed the change-kvcache-config branch 2 times, most recently from ff511c3 to 712c60b Compare July 9, 2025 00:28
@Superjomn
Copy link
Collaborator Author

/bot run

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11358 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11358 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #8406 completed with status: 'FAILURE'

@Superjomn Superjomn force-pushed the change-kvcache-config branch from 712c60b to d37c6f9 Compare July 9, 2025 05:35
@Superjomn
Copy link
Collaborator Author

/bot run --stage-list "A30-PyTorch-2"

@Superjomn Superjomn force-pushed the change-kvcache-config branch from d37c6f9 to 4c7f2e6 Compare July 9, 2025 05:51
@Superjomn
Copy link
Collaborator Author

/bot run --stage-list "A30-PyTorch-2"

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11400 [ run ] triggered by Bot

@Superjomn Superjomn requested a review from FrankD412 July 9, 2025 06:18
Signed-off-by: Superjomn <328693+Superjomn@users.noreply.github.com>
@Superjomn Superjomn force-pushed the change-kvcache-config branch from 4c7f2e6 to 5634f17 Compare July 9, 2025 07:14
@Superjomn
Copy link
Collaborator Author

/bot run

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11406 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11400 [ run ] completed with state ABORTED

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11406 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #8437 completed with status: 'FAILURE'

@Superjomn
Copy link
Collaborator Author

/bot run --disable-fail-fast

@tensorrt-cicd
Copy link
Collaborator

PR_Github #11442 [ run ] triggered by Bot

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants