code wiki / (root) / nx_llm_sched_gate.nx

nx_llm_sched_gate.nx

buildroot/runtime/nx_llm_sched_gate.nx

8978 B228 linesdepth 16pulls 67 transitivereach 0 importersview sourcekind gate/prooftopic llm
docsdependenciesstructsconstsfunctions

about

nx_llm_sched_gate.nx -- MEASURED gate for the CONTINUOUS-BATCHING scheduler (nx_llm_sched) on the REAL model. STAGGER 4 requests admitted at different times (2 up front, 1 after round 2, 1 after round 4) -- requests JOIN the running batch mid-flight (the continuous-batching property) INVARIANT each request's bytes BIT-IDENTICAL to its SOLO sequential run (same seed) -- output independent of co-tenants and admission timing (the fundamental serving correctness) HYGIENE pool all-free after releases license_tier: ORIGINAL expect_exit: 0

dependencies 27 imports · 0 importers

nx_syscalls.nx nx_tier.nx nx_le.nx nx_bpe.nx nx_gguf.nx nx_gguf_load.nx nx_gguf_meta.nx nx_f32.nx nx_f32_kv_cache.nx nx_f32_lazy_weight.nx nx_llm_sched_gate.nx

diagram shows first 10 each side; +17 more imports, +0 more importers in the complete lists below.

imports: nx_syscalls.nxnx_tier.nxnx_le.nxnx_bpe.nxnx_gguf.nxnx_gguf_load.nxnx_gguf_meta.nxnx_f32.nxnx_f32_kv_cache.nxnx_f32_lazy_weight.nxnx_f32_llama_block.nxnx_f32_llama_block_v4.nxnx_f32_llama_stack_v4.nxnx_f32_llama_layer_lazy_load.nxnx_f32_llm.nxnx_f32_llm_v4.nxnx_f32_llm_read_dims.nxnx_f32_bpe_load.nxnx_f32_llm_special_tokens.nxnx_f32_sampler.nxnx_prng.nxnx_reasoning.nxnx_kvcache.nxnx_f32_attn_paged.nxnx_f32_llama_v4p.nxnx_f32_llama_v4b.nxnx_llm_sched.nx

imported by: nobody (leaf or entry point)

call flow from main pre-order; caps 40 nodes / depth 6 declared; ↻ = already shown

main sys_mmap sys_read_file sys_openat_rd sys_lseek sys_mmap ↻ sys_read sys_close nx_gguf_parse nx_le_read_u32 nx_le_read_u64 nx_le_read_u32 ↻ _gguf_skip_value nx_le_read_u64 ↻ nx_le_read_u32 ↻ _gguf_skip_value ↻ sys_mmap ↻ nx_f32_llama_model_alloc sys_mmap ↻ nx_f32_llm_read_dims_from_ sys_mmap ↻ nx_gguf_meta_find nx_le_read_u64 ↻ nx_le_read_u32 ↻ _gmeta_bytes_equal nx_gguf_skip_value _gguf_skip_value ↻ nx_gguf_meta_read_string_l nx_le_read_u64 ↻ nx_gguf_meta_read_string_p _fld_read_arch_u32 sys_mmap ↻ _fld_concat_key nx_gguf_meta_find ↻ nx_gguf_meta_read_u32 nx_le_read_u32 ↻ _fld_read_vocab_size_from_ sys_mmap ↻ nx_gguf_find_tensor nx_gguf_tensor_at

structs

none

consts

42const SG_MAXNEW: nx_int = 6

functions

44func sg2_w(s: *u8) -> i64 { var n: i64 = 0; while s[n] != (0 as u8) { n = n + 1 } sys_write(1, s, n); return 0 }
called by 2: sg2_wnmain calls 1: sys_write
45func sg2_wn(v: i64) -> i64
called by 1: main calls 3: sg2_wsys_mmapsys_write
58func sg2_beq(a: *u8, b: *u8, n: i64) -> i64
called by 1: main
66func sg2_solo(rc: *NxReasonCfg, pool: *NxPagedPool, toks: *i64, nt: nx_int,
called by 1: main calls 8: nx_pkv_seq_newsys_mmapnx_f32_llm_forward_v4pnx_prng_initnx_f32_sampler_sample_top_knx_bpe_decode_bytelevel+2
112func main() -> i64
calls 22: sys_mmapsys_read_filenx_gguf_parsenx_f32_llama_model_allocnx_f32_llm_read_dims_from_ggufnx_f32_llm_load_weights_v4_from_gguf+16