leavelet
864b1c808e
On a radix miss, reload the evicted-but-L3-durable suffix from disk into L2 + insert a fresh radix node, holding the triggering request until then (mirrors the storage-prefetch hold, which is disabled under CP). The reloaded prefix re-enters the radix as a normal L2 node, so the existing load_back serves L2->L1 -- no new device path. Design (docs_internal/cp_hicache_l3_phase3_impl_design.md §9, refined for minimal sync): - ENTRY (scheduler _prefetch_kvcache, enable_cp_l3 branch): mark a miss-with-suffix as a reload candidate (rank-uniform: requests are broadcast). cp_l3_reload_lookup computes the suffix's full-page content hashes -- the SAME SHA chain spill wrote (compute_node_hash_values). - RESERVE + ADMIT fold into the existing per-tick _drain_l3_control_queues MINs -- ZERO new collectives: the candidates' per-rank exists_prefix counts ride the qsize MIN vector -> a rank-uniform agreed count C (this reconciles the async-LMDB read-skew); reload-ack oks ride the durable ok-AND. Reserve is then collective-free (deterministic evict-to-fit mirroring _reserve_write_cp_shared_l2_evict_to_fit); the request waits non-blocking in waiting_queue (check_cp_l3_reload_progress skip) and is admitted + re-matched the SAME tick the reload lands (check_hicache_events runs before batch formation). CP-aware insert only on a still-clean attach (else abort+recompute). Capped (cp_l3_reload_max_inflight) + content-key deduped (piggyback). EAGLE/bigram (opus-studied, PROVABLY correct -- not deferred): the radix key is bigram-converted over the whole request, and convert_to_bigram_key(fill_ids[M:]) == bigrams(fill_ids)[M:] exactly (the boundary bigram belongs to the matched prefix), so the bigram-converted suffix hashes reproduce node.hash_value. Verified vs source + a new slice-identity regression test. Opus review (FIX-THEN-SHIP, no CRITICAL; the 4 crash surfaces -- MIN-vector shape, read-skew, attach-anchor, placement-digest -- traced clean) + independently re-verified; fixes folded: - M1: cp_l3_release_request clears reload state on request abort/timeout/preempt (no leak). - H1: rank-uniform per-op TTL bound releases a held request if its reload op never acks (the default SGLANG_REQ_WAITING_TIMEOUT is off); the reservation frees safely at the (late) ack. - L1: fail-soft anchor guard at insert (re-derive the first suffix hash from lhn's live hash). - empty-waiters reload aborts (frees L2) instead of inserting an unrequested node. Imports hoisted to top-level (get_hash_str -- hicache_storage is a leaf module, no cycle; convert_to_bigram_key already top-level), no inline imports. py_compile clean; 54 L3 unit tests green. Reload triggering (submit_reload/exists_prefix/free_object) now wired (was write-only). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Registered Tests
Tests under this directory are auto-discovered by run_suite.py via CI registration decorators.
Where Should I Put My New Test?
No server / engine launch required
| What you're testing | Directory | Requires |
|---|---|---|
| Component logic in isolation (cache, scheduler, config, parser, etc.) | unit/<module>/ |
CPU or GPU |
| CUDA kernel correctness | kernels/ |
GPU |
Server / engine launch required (E2E)
| What you're testing | Directory | Requires |
|---|---|---|
| Model inference correctness | models/, 4-gpu-models/, 8-gpu-models/ |
GPU |
| Feature-specific (OpenAI API, LoRA, speculative, distributed, VLM, etc.) | openai_server/, lora/, spec/, distributed/, ... |
GPU |
| Benchmarks (performance, accuracy, stress) | benchmark/ |
GPU |
| Platform-specific | amd/, ascend/ |
Vendor GPU |
See unit/README.md for unit test conventions.