Files
sglang/python
leavelet ed42a6dcbc Honor x-request-id as rid + env-gated NSA tensor dump for L1-vs-L2 reload diff
- http_server: chat + generate endpoints read x-request-id header into rid (the
  Rust PD router drops the client body rid but forwards the header), so the
  client id reaches forward_batch.rids for exact cross-send join.
- cp_hicache_trace.dump_tensors + SGLANG_NSA_DUMP_DIR: torch.save q/composed-KV/
  selection/attn_out at layer 0 for rids starting 'dump-' (extend forwards), to
  diff L1-hit vs L2-reload by relative error (beats fp-nondeterminism that
  defeats binary hashes). Wired into nsa_backend flashmla_sparse path.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-17 14:05:54 +00:00
..
2026-06-10 05:54:43 +08:00