Repository navigation
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Add ReplaySSM state modes for GDN and KDA with explicit native, compact, and replay state strategies. The change separates Conv/SSM state ownership, commit/acceptance handling, PD transfer regions, replay kernels, autotune bounds, and state-mode validation while preserving existing non-replay modes.
This branch is a clean single commit based directly on
upstream/main; it intentionally excludes the merge history from the superseded ReplaySSM branch.Validation
python -m compileall -q lightllm unit_tests testgit diff --checkpre-commit run --files <all changed files>(Black and Flake8)pytest -q unit_tests/common/test_linear_state_contract.py unit_tests/server/test_ssm_state_mode.py(115 passed)The Triton kernel suite requires a compatible CUDA/Triton runtime and is not represented as a passing result here; GPU execution remains an upstream CI/H100 validation item.