Repository navigation
Preserve native DeepSeek snapshots when admission is rejected - #54
Draft
beastllama wants to merge 1 commit into
Draft
beastllama wants to merge 1 commit into
beastllama wants to merge 1 commit into
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Root cause: prompt-cache lookup evicts snapshots to reserve future saves before the backend accepts placement, even for native DeepSeek-V4.1 whose drops reach followers immediately. Evidence: the CPU regression on
e100ee8consumes aContextFullrefusal without another prefill, leaves the request queued beside the existing stream, and nevertheless evicts one reusable snapshot.This change lets native DSV41 defer reservation to the existing budget-enforcing
keepBornpath.Rules.reserve_before_passdefaults totrue; native DSV41 alone opts out throughLoaded.snaps. Families whose peers receive drops with the next request retain their existing reservation behavior. No cache budget, admission capacity, arithmetic or wire protocol changes.The deterministic LaneHost regression checks that refused placement preserves both warm snapshots. Its success control checks that admitted work still evicts exactly one snapshot, saves its mark, and stays at the same 128-byte fixture budget.
Validation with the repository-pinned Zig 0.17.0, CPU only:
expected 0, found 1evictions).zig fmt --check, andgit diff --checkpass.Reproduction commands from the repository root (
zigmust be 0.17.0):Local limitations: the broader engine API suite reached 42/179 before an integer-overflow panic in the existing
prompt_imprint_fault_testallocation-failure test; that same panic was independently reproduced on an unmodifiede100ee8archive. The native build runner failed before source compilation withchecking cache failed: manifest_stat Unexpectedunder the available WSL environment, including with isolated Linux source/cache directories. A complete native build and full host suite are therefore not claimed passed. GPU/model qualification was not run for this public-branch port; no serving process was changed and no performance gain is claimed by these CPU fixtures.