`test_bag_values_match_server_args` asserted `bag == field`. That holds today
only because construction resolves in place; step 12 keeps the record raw, and
the plan doc calls this test out as one that becomes **false by design** for
every field resolution fills in.
Rewritten against the resolved projection, which is the half that survives: the
bag carries what resolution produced. The `bag == field` assertion stays as one
line at the end, labelled as the tripwire -- when it starts failing for a
resolution-written leaf, the flip has landed and the bag is the only place the
effective value lives.
The reference is an independent resolution of the same raw input (a fresh,
never-published record) rather than `resolved_server_args_dict()`, which reads
`vars(server_args)` back and therefore only restates the published instance.
And the record goes through the real pipeline on a real mini config, published
through `publish()`: the dummy-model path returns at the dummy boundary with
every sampled leaf still raw, so the old comparison was raw==raw and vacuous
(both Codex catches). Reproducibility (#34094) licenses the sibling as a
stand-in for the pipeline output.
The sample admits only leaves resolution writes on this input on both CI
device shapes (attention_backend, page_size, chunked_prefill_size,
mem_fraction_static), and the raw-differs guard asserts it per leaf -- a
default-count threshold let supplied inputs like `model_path` (no dataclass
default, so any path "differs") stand in for resolution work. Passthrough
leaves (host, hicache_ratio, moe_runner_backend, model_path) move to a
separate projection smoke that claims only what it checks: publish projected
an unchanged field into its namespace. Between the two resolutions the test
restores environ and the EnvField none-flags, so the sibling resolves the
same pristine input rather than the first resolution's leftovers.
And the class runs its body exactly once, like the other dual-resolve
harnesses: a CI retry re-enters after the first attempt leaked process
state, which is the hazard the pristine snapshot exists to rule out.
docs(skill): a supplied-instance read is not automatically safe
The whole-object rule said "keep the supplied-instance contract; don't rewrite
the parameter reads unless the field is runtime-mutated". That is the right rule
for the *object* and the wrong stopping point for the *field*: after step 12 the
record carries the user's raw input, so `server_args.page_size` inside a
runner-owned constructor reads the CLI default rather than the effective value.
The rule now names that second case as step-12 debt with a guard attached
(`test_supplied_instance_exposure_ratchet.py` fails on a new pair, so the
decision is made when the read is written), and names the two shapes that stay
parameter-form on purpose: a helper the resolution pipeline calls with a
`resolved_view`, and a factory whose contract is "build X from the record you
are handed".
A callee that takes `server_args` keeps the supplied-instance contract, so no
ratchet counts its reads -- and that is right for the *object*. What it does not
cover is what the object will carry after step 12: the instance stays at the
user's raw input, so a callee reading a field resolution fills in starts seeing
the CLI default instead of the effective value.
Measured, not guessed: **297 distinct file/field pairs read one of the 125
fields resolution can write** -- what remains after the earlier members'
conversions (the census found 314 across the package; the page_size,
chunked_prefill_size and graph/limit families were converted by the members
below this one, so the list lands at the remaining debt with no churn). The
census counts three spellings of the read: `server_args.field` off the
parameter, `getattr(server_args, "field", default)` with a literal name, and
the *parked* form -- `self.x = server_args` in a method that takes the
parameter, read as `self.x.field` anywhere in the class. Parking under a
different object, a container, or a computed name stays invisible, like in
every census of this family.
The written-field set is derived in-test from resolved configs against the
dataclass defaults -- the same matrix the context repo's audit tool uses -- and
the union is only as complete as the matrix: fields a matrix entry passes in
are excluded, so each entry must let resolution make the decision the entry is
about. The DWDP shape is the loudest example: `_handle_dwdp` writes `dp_size`,
`enable_dp_attention`, `ep_size` and friends itself, and without a
`{tp_size: 2, dwdp_size: 2}` entry the whole DP/EP topology family (37 pairs
across the launcher, the controllers, the tokenizer and the spec workers)
never entered the written set at all. Multi-item scoring is the same shape in
miniature -- `_handle_multi_item_scoring` writes `disable_radix_cache` itself,
and the `{enable_mis, attention_backend=flashinfer}` entry (the backend is
passed because the handler asserts rather than switches) pins the radix-cache
builder and its friends.
The written set carries **may-write semantics**, statically collected from
every mechanism that can put a resolved value on the record, unioned with the
matrix (construct-and-diff still catches value-level writes statics cannot
name). The enumeration approach kept losing to review -- the DFLASH hook hole,
then the declarative registry (`MODEL_OVERRIDES` forces dtype for two arches
through a setattr applier no assignment scan sees), then the deprecated-alias
loop that writes through a name tuple -- because each round found one more
*mechanism*, not one more field. So all of them are collected now: hook
assignments under `arg_groups/`, the record's own method assignments (the
mooncake layout rewrite, the deepseek-EP mode defaults, the seed fill that
only fires when the caller did NOT supply one -- which construct-and-diff can
never see, since measuring requires supplying), the declarative override
registry, the alias-normalization tuple (drift-guarded), and the
late-resolution keywords. A statically-collected write site that can never
fire is a dead branch to delete upstream, not a reason to shrink the census
(maintainer's ruling). And the pin is split by host: `_EXPOSED` is asserted everywhere,
`_EXPOSED_CUDA_ONLY` (empty today) is where a capability-gated write's
readers go -- one shared exact list cannot hold such a pair at all, since
pinning it fails CPU as "gone" and omitting it fails CUDA as "new".
The resolve-once shape also undercounts on axes a construction call never
sees (each of these was a review catch, and each added pinned families):
resolution branches on the *environment*, so every matrix entry resolves under
the plain env and under the CI shape (`SGLANG_IS_IN_CI`), with the pristine
process state (environ plus the `EnvField` descriptor flags) restored between
entries -- that is where `soft_watchdog_timeout`'s four readers come from. Some
fields resolve *late*, at validation rather than construction
(`declare_late_resolution`): those writers are collected statically by keyword
-- `lora_paths`, `reasoning_parser`, `tool_call_parser` -- with the one dynamic
`**detected` site spelled out in a table guarded against drift. Fields holding
only a `default_factory` are materialized rather than skipped, and
`tokenizer_path` / `served_model_name` -- always filled from `model_path` --
leave the passed-inputs exemption and pin their twelve readers. A module the
census cannot parse fails the test instead of shrinking it, which immediately
caught a BOM-carrying file every previous census had silently skipped (the
scans read `utf-8-sig` now). And the test registers on the CUDA runner besides
the CPU suite, because capability-gated writes only open on real hardware;
AMD is intentionally not registered -- an exact pin cannot be verified from
any pinning host -- with the reasoning in the header.
**A second axis is already wrong today**, independent of step 12. Some config is
decided after publish and recorded with `get_context().override(...)` -- elastic
EP resizing `ep_size`, a weight update rewriting `model_path` / `load_format`,
HiCache attach naming a storage backend, adaptive speculative decoding moving
`speculative_num_steps`. That write reaches the bags and never the record, so a
supplied-instance read of one of those fields answers with the startup value from
the moment the override lands. **73 pairs over 13 fields** are in that position -- including the overrides
that arrive as `**kwargs`: the collector statically resolves dict-literal
expansions (the HiCache attach shape, whose write/read pairs on
`hicache_write_policy` / `hicache_storage_prefetch_policy` were invisible
before) and fails loudly on anything it cannot resolve, with
`update_server_args` exempted by name because its key set is the API's
caller's, not this file's.
Whether each is a defect depends on ordering -- a value copied at construction,
before any override, is fine -- so the axis is pinned as a measurement with the
same growth guard, not as a list of bugs.
One of them *was* a defect and is fixed at the base of this stack: the
linear-attn dispatch table rebuilt itself from the record after the SM100 GDN
prefill decision had been recorded in the bag, so a second runner's rebuild
dropped it. That choice is a per-runner stamp now and is not recorded
process-wide at all, which is why neither the read nor the field appears on this
axis.
The list is pinned both ways, on both axes. A new pair fails, because the moment
to decide where a resolved value comes from is when the read is written, not
during the flip; a disappeared pair fails too, naming the entry to delete, so the
registry stays a measurement rather than a memory of one. Both axes
reverse-verified: a new read of a written field is reported by file and field.
Per-field dispositions live in the plan doc; several of these are "should this
callee take a config at all?", which is a design call rather than a sweep.
test(step-12): tripwire on the EPD guard that a raw record would silence
`_reject_missing_dispatched_encoder_embedding` is one of the two reads the
step-12 audit calls a blocker: it keys on `encoder_transfer_backend`, a field
resolution fills in, off a handed record. Today that record carries the
resolved value; after the flip it stays at the argument default `"auto"`
(`ENCODER_TRANSFER_BACKEND_CHOICES[0]`) for every auto-resolved launch and the
503 stops firing -- a guard that goes quiet, which no existing case notices.
The tripwire resolves a real language-only Kimi-K3 TP2 launch (a mini config,
the shape whose auto pick is `"zmq_to_tokenizer"`) and asserts the guard
rejects with the record resolution produced. A fixed double cannot trip on the
flip -- it would keep handing the guard the resolved value by construction --
so the record has to come from resolution itself: when step 12 lands, this
same launch hands the guard `"auto"`, the rejection silently stops, and this
test fails, which is exactly the signal that this reader needs the resolved
value from somewhere else (the per-engine overlay or the bag).
The launch pins `mamba_radix_cache_strategy=no_buffer` (+ the overlap-off it
requires): resolution's hybrid state-cache sizing branches on the host device
and asserts a GPU stack for extra_buffer, which a CPU CI runner does not have,
while the guard under test reads a field independent of that branch. The case
restores env *and* the EnvField descriptor flags -- a real resolution leaves
state os.environ does not carry.
config: the speculative workers take page_size from the bags
Seven worker constructors stored `self.page_size = server_args.page_size` off
the handed record. They all run after publish and all keep a copy of a
process-level value, which is the first row of the plan doc's supplied-instance
disposition table -- so they read `get_schedule().page_size`, and a post-publish
override now reaches them like it reaches every other consumer.
The supplied-instance census named the seven pairs; the exposure ratchet in the
next member pins what remains after this batch of conversions.
config: the post-publish chunked_prefill_size consumers read the bags
Four of the ten supplied-instance `chunked_prefill_size` reads are plain
post-publish consumers -- the EPLB recorder's buffer sizing, the deep-gemm
compile warmup (five reads), the KV-cache builder's effective size, and the
ngram embedding manager's assert. All are reached from runner init, so they read
`get_schedule()`.
Two are deliberately left: `create_kt_config_from_server_args` builds a config
*from a supplied record* by name and contract, and `CanaryLaunchCapacities.from_args`
is the same shape. Converting those would change what the function is, not where
it reads -- the plan doc's disposition table says so per field.
config: the remaining post-publish graph/limit consumers read the bags
Three more of the census's supplied-instance debts are plain post-publish reads: the dspark worker's
cuda-graph decode sizes, the dspark planner's SPS table bound
(`max_running_requests`), and the LoRA manager's cuda-graph moe buffers. The
dspark worker is the clearest of them -- it already read
`get_exec().graph.cuda_graph_config.decode.bs` thirty lines below the instance
read, so the file disagreed with itself about where the same value comes from.
Left where the function's contract is "build a config from the record you are
handed" rather than "read this process's config":
`create_kt_config_from_server_args`, `DllmConfig.from_server_args`,
`CanaryLaunchCapacities.from_args`, `build_compilation_config`. Changing those
would change what the function is.
config: the runner, scheduler and offload manager take page_size from the bags
The same `self.page_size = server_args.page_size` shape as the speculative
workers, in the three remaining process-owned constructors: `ModelRunner`,
`Scheduler`, and the decode-side KV offload manager. The scheduler process
publishes before any of them run. The one path that did not is `ModelRunner`
constructed standalone -- `python -m sglang.benchmark.one_batch` and the manual
runner tests build it with no prior publish, and the constructor's own publish
sat below this read -- so that publish moves above the constructor's first bag
read instead of leaving a window where the runner half-exists unpublished.
Left where the read belongs to something else: `utils/common`'s predicates are
called only from the resolution pipeline with a `resolved_view`,
`allocation_sizing` takes the config its callers supply by contract, and
`CudaVmmFeatureTransport` is tokenizer-owned -- one per tokenizer worker, which
is the per-instance boundary.
The conversion left the offload manager parking a record it no longer
reads; the parked copy goes with the read (the constructor parameter stays
-- its hicache sizing still reads it directly).
The previous batch counted `self.server_args.X` and called the runner surface
done. It was not: the same read spelled through a local alias --
`server_args = model_runner.server_args` (or `sa = kvc.server_args`, `args = ...`)
followed by `server_args.leaf` -- is the same process-global read wearing a
local name, and the AST census counts **57 of them** across eleven files that
the grep never saw. Census per function, following the alias.
52 were leaves and go to their bag (`spec` 11, `schedule` 9, `memory` 7,
`exec.graph` 5, `exec.moe` 5, `parallel` 4, `disagg` 4, `model` 3,
`exec.mamba` 2, `exec.overlap` 2). Five were not leaves:
three derived members on the eager runner --
`max_speculative_num_draft_tokens` and `enable_mamba_extra_buffer` already had
accessors, and `max_prefill_buffer_tokens` gets one (all its inputs are `schedule`
leaves plus the configured PP size, so it derives from the bags and follows a
post-publish override; `TestDerivedPredicatesAgreeAcrossTiers` pins it against
the member over a 48-case matrix) -- plus `get_attention_backends()`, which the
same commit routes through `attention_backends()`, and a dict that merely shares
the name (`server_args_dict.items`). That dict is the one read left behind.
`build_attention_backends` also stops resolving the pair from the record: it
runs after publish, so it asks `attention_backends()` like every other consumer.
The draft override on the runner still wins first.
`dispatch_event_loop`'s three PP checks read the *configured* PP size, not the
live topology: the MLX runner stub never initializes torch.distributed, so the
live property asserts before the MLX event loop can start (a Codex catch). The
configured leaf answers the same value wherever the live groups exist.
`flashinfer_gdn_prefill_default`'s guard is the one read here that asks what the
*operator* named rather than what the config resolved to, and the bag leaf now
answers exactly that: the per-runner auto-default is stamped on the runner and
deliberately never recorded process-wide, so nothing writes that leaf after
launch and reading it back cannot mistake another runner's default for a flag.
Three test doubles injected a `SimpleNamespace`/`MagicMock` record for exactly
these reads and now publish instead (pool configurator, cache registry, GDN
prefill policy) -- the fixture publishes what the case configures and hands the
published instance to the whole-object contracts that still take one.
The functions this sweep partially converted stop mixing sources (review
catches): the flash-attention constructor's remaining seed reads
(`speculative_eagle_topk`, `speculative_algorithm`, both deterministic gates)
read their bags next to the leaves already converted;
`_should_disable_scheduler_metadata_precompute` reads the parallel config
leaves itself instead of taking the record (its alias binding was the last
use); and the autotune gates (`disable_flashinfer_autotune`, deterministic,
`flashinfer_autotune_skip_ops`) join the moe leaves the same function already
reads from the bags. The pool-configurator fixture drops a parameter nothing
published or read.
`_is_dsa_active` asked `getattr(server_args, "_is_dsa_model_arch", False)`, and
that name has never existed on `ServerArgs` -- it arrived as a placeholder with
the CP strategy abstractions (#27313), so the getattr default has always decided
the predicate. A dynamic read of a name nothing sets is the one shape the config
census cannot follow, and it looked like a live decision while being dead.
Spelled as the constant it evaluates to, with the placeholder written down: what
it should ask (whether this process runs a DSA model arch) is the CP path's
call, and its only consumer, `ContextParallelStrategy.per_layer_attn_cp_comm`,
has no readers yet.
That was the sole entry in the read ratchet's `_INERT_DYNAMIC_READS`, so the
exemption list is gone with it -- there is no way to exempt a read from the
baselines any more, which is the invariant worth having. The `counted()`
indirection it existed for goes too (verified the three shapes it guarded still
report: direct, `getattr`, and an attribute-parked alias).
`--attention-backend` is one field of three: a launch that sets only
`--prefill-attention-backend` or `--decode-attention-backend` leaves the base
field at `None`. Seven decisions read that base field alone and therefore
answered from a field the operator never set. `attention_backends()` is the
pair with the base-field fallback already applied, so each site now asks it for
the half it actually needs:
- `inkling_common/attn` assembles backend-specific kwargs (rel_bias / score
mods) and gates its fused prologue; the backend those describe is the one
`self.attn` dispatches to, so `serving_attention_backend()` selects the pair
member by `forward_batch.forward_mode`, mirroring
`HybridAttnBackend._select_backend` exactly -- draft-extend routes through
the prefill branch like the dispatcher does -- and preferring the
runner-stamped pair, so a draft runner answers with its own backend. That
preference only works if every backend that can enter a ForwardContext
carries the stamp, so `DraftBackendFactory._create_backend` now stamps its
products with the backend it resolved (draft override first), and the
draft-extend conv-sidecar wrapper copies the wrapped backend's stamp -- the
replacement backends the spec workers install had no stamp at all and fell
back to the target's configured pair.
- The chunked-prefix-cache gate is a *prefill* feature -> prefill half. Reading
the base field switched the feature off for every prefill-only configuration.
- `init_deterministic_inference_config` maps *prefill* knobs
(SPLIT_TILE / PREFILL_TRUNCATION_ALIGN) -> prefill half; the map missed and
left truncation unset.
- `two_batch_overlap` computes extend positions -> prefill half.
- mrope's interleaved-rope kernel runs in both phases -> both halves must
support triton. This one is not conservative when it misreads:
`support_triton(None)` answers **True**, so a `--prefill-attention-backend
torch_native` launch took the triton path.
- The req-to-token writer has one caller, `alloc_for_extend` -> prefill half;
its fallback pays several `.item()` syncs per request, so gating it on the
decode half too would send every extend of a mixed launch through the slow
path. `get_last_loc` (the spec-decode allocator's helper) keeps the
both-halves reading: verify tokens are served by either half depending on
`speculative_attention_mode`.
- The flashinfer version floor is a guard; it never fired for a launch that
pinned flashinfer through a split field.
One more site the census found is not converted here: `gpt_oss` derives its
`sinks` parameter dtype from the backend, and a single parameter dtype cannot
serve a split pair (FA4 asserts bfloat16, trtllm_mha consumes float32), so
that one is a behaviour question rather than a config-source one and is fixed
in its own PR.
`test_split_attention_backend_decisions.py` pins the callable decisions by
calling them under a split-only publish, and pins the remaining ones
statically -- the file/why map fails if any of them goes back to the base field
(reverse-verified). It also asserts the `support_triton(None) is True` trap the
sweep exists for.
The stamp comes from the constructor, not the request: every factory leaf
answers ("effective_name", backend), because several map entries do not build
what their key says -- cutedsl_mla draft-extend builds the trtllm-mla backend,
"nsa" is a deprecated alias building dsa, and the hybrid-linear entries pick
fa3/intel_amx/triton by host, which no static rename table can express (a
review catch: on Blackwell the alias stamp reached Inkling's per-forward
kwargs assembly, which asserts a concrete kernel name, and crashed the first
draft-extend forward). The stamping is pinned by unit tests, not only by a
spec e2e: removing the child-stamping loop, stamping an alias from a leaf, or
dropping the wrapper copy goes red (reverse-verified), and a static guard
walks the factory source asserting no leaf answers an alias name. The child loop states its contract explicitly --
`create_decode_backend` passes `stamps_children=True` because its products
are per-step containers by construction, so a container without
`attn_backends` raises instead of being silently skipped by a defensive
probe. The `_version` invalidation names its contract (autograd's in-place
counter: private, chosen because it is the only per-tensor signal that ticks
on copy_-style updates; removal fails loudly). The version-floor guard's file
joins the pair-reader ratchet, and the one runner-seed chain read sharing the
backend's __init__ (`speculative_eagle_topk`) reads the spec bag.
Six reads were left on `self.server_args` outside the per-instance boundary the
plan reserves for the tokenizer-manager family, and each had a different reason
to be there:
- `scheduler.process_input_requests` (`mm_feature_transport`) and
`BaseSpecWorker._build_hicache_draft_plan` (`enable_hierarchical_cache`) are
plain leaves -> `get_mm()` / `get_memory()`.
- `DraftBackendFactory._create_backend` read the split backend through a
*runtime-computed name* (`getattr(self.server_args, backend_name)`) and then
fell back to the base field by hand -- the census's documented blind spot.
The two names it can be handed are exactly the pair `attention_backends()`
returns with that fallback already applied, so it reads the pair and indexes
it. The draft runner's own stamp still wins when it has one.
- `remote_instance_weight_loader_use_transfer_engine` and
`pre_capture_activation_reserve_mb` are derived members. Both are computed
from published leaves only, so both get a named accessor that derives from
the bags (and therefore follows a post-publish override).
The first of those two has all its inputs in one bag, so it follows the
established shape: one `*_of(cfg)` helper in `arg_groups/overrides.py`, the
`ServerArgs` member delegating to it, and the accessor calling it on
`get_model()`. `modelexpress_transport_of` splits out the JSON parse both
sides need. The second spans four bags plus the configured parallel sizes, so
it exists twice like the mamba pair -- and `TestDerivedPredicatesAgreeAcrossTiers`
now pins both new pairs equal over their input matrices (92 subtests).
`self.server_args.X` outside the tokenizer-manager family: 11 -> 5, and the
five that remain are the documented ones (the encode server's own record, the
nixl connector's rank arithmetic, `GrammarManager`'s handed instance).
The post-capture headroom path calls the same bag-backed
`pre_capture_activation_reserve_mb` accessor the configurator uses -- the
accessor advertises override-following, and a reserve that reads the record
while its sibling reads the bags can disagree after a post-publish override.
And the conversions' orphans go with them: `RemoteInstanceWeightTransporter`
kept a `server_args` field nothing reads, and `DraftBackendFactory` parked a
record it no longer consults -- both drop the parameter, and the four factory
call sites stop threading one.
Two problems in the same family as #33312 (a per-runner decision that one
participant answered differently), one fixed here and one guarded.
**The linear-attn kernel backends were process-wide.** `attn_backend_wrapper`
rebuilt a module-level dict once per runner, from the handed record plus a local
`prefill_default`. Two things follow, and both are wrong:
- **A draft could not hold a different choice than its target.** Only the runner
whose model is GDN gets the SM100 FlashInfer prefill default; the operator's
explicit flag belongs to the launch. The full-attention backends already model
this correctly -- the runner stamps `prefill_attention_backend_str` /
`decode_attention_backend_str` and its backend objects are built from the
stamp. Linear attn had no stamp at all.
- **The second rebuild replaced the first one's choice.** The default was also
recorded into the process-wide config, which the record does not see, so a
runner rebuilding without a default of its own resolved `prefill` back to the
base backend -- silently swapping the kernel the earlier runner selected.
Demonstrated in-process before this change: table `FLASHINFER`, then `TRITON`.
`resolve_linear_attn_backends(prefill_default=None)` returns a frozen
`LinearAttnBackends(decode, prefill, verify)` from the published `exec.mamba`
leaves; the wrapper stamps it as `runner.linear_attn_backends` before building
the backends that read it; and the three consumers (GDN, KDA, Ascend GDN) read it
off the runner they are built for. Each already took `model_runner` and cached
the result on itself, so the value now simply comes from the right place. A
backend built outside that path has no stamp and raises on the attribute, the way
the full-attention strings do -- no silent fallback to hide the wiring mistake.
The recording goes away with it. A per-runner choice in the process-wide config
has no meaning the second runner can read correctly: the leaf is how the gate
asks "did the operator name a backend", so a recorded default reads back as an
operator flag and the next runner declines its own. The leaf now keeps meaning
what was asked for at launch, and the effective choice lives in the stamp (and
in the log line the gate already emits).
Precedence is unchanged: the resolver takes the default as an argument and an
explicit `--linear-attn-prefill-backend` wins over it, with the gate declining
early so it neither probes the device nor logs.
**A draft entry class must answer the loader exactly when its target does.** The
loader asks the entry class it instantiates for the shared-experts-fusion
decision, and a draft is its own entry class. When the target family carries
auto-disable conditions and the draft's class does not expose them, the loader
installs one decision for each and the draft's weights are laid out for the wrong
one. That shipped: the DSV4 DSpark draft skipped its bundled shared-expert
tensors until #33312 gave it the gate, costing accept length 5.60 -> 2.05.
`test_fusion_gate_coverage.py` walks the same registry but asks whether an entry
class *touches* the decision -- reads the flag, names a gated class. That catches
a class once it already consumes the decision; it could not catch one that should
consume it and does not, which is what the DSpark class looked like (it built the
family's *layer* classes, so the flag reader lived in another module and its own
source named no gated class). `test_draft_entry_hook_parity.py` asks the
invariant directly: presence parity between a draft entry class and the target it
is named after. Identity is deliberately not required -- the Qwen3.5 MTP
delegates with adapted arguments (unwrapping `text_config`, using the MTP
quantization config), which is right -- and weight-name maps are out of scope,
since a draft's checkpoint has its own names.
Reverse-verified against the original defect: with the DSpark gate removed the
case names the pair and the side that is missing it; with #33312 in place it
passes.
`build_draft_tp_worker` built a `ServerArgs` variant whose only job was to make
four config reads answer with the draft's backend instead of the target's, and
published it for the duration of the build so the bags agreed. The backend is a
per-runner fact — target and draft coexist in one process — so it moves onto the
runner, and the variant and the construction-time publish both go away.
`ModelRunner` takes `draft_attention_backend` and resolves the runner's effective
value once (`resolve_draft_attention_backend`: the algorithm's resolved backend,
else `--speculative-draft-attention-backend`, else None for a target runner);
`TpModelWorker` threads it to both runner constructions.
`resolve_attention_backend_strs` reads it off the runner, and `ModelRunner`
stamps the resolved pair *before* building backends so a backend can read it
while it constructs — which is what the FlashInfer KV-access check needs now that
it no longer asks the config. `configure_kv_cache_dtype` and the draft backend
factory read the runner too.
One latent bug falls out: the non-hybrid branch of the backend build ignored the
resolved pair and re-read `server_args.attention_backend`, which is why the
variant had to set that field as well as the split pair. It now uses the value
that was resolved for the runner.
`draft_server_args_overrides` and the `preserve_config()` publish switch are
deleted; with them goes the last production `ServerArgs.derive` outside
pre-publish config building, and the last construction-time publish. The
chunked-prefix gate the target resolved simply stays in the bags, since nothing
re-projects them.
The v2 spec workers got a published `ServerArgs` copy carrying two values: the
target's context length and `--speculative-draft-load-format`. Neither is a
process-wide config change — each is consumed by exactly one constructor — so
the copy, the publish switch around the draft build, and the replay of the
target's resolved overrides onto it all go away, and the values travel to the
runner that owns them:
- **Context length.** `TpModelWorker` already takes it (`context_length=None`
keeps `server_args.context_length`); the four v2 draft workers and
`build_draft_tp_worker` pass the target's, which every one of them has in
scope as `target_worker` / `target_model_config`.
- **Load format.** `ModelRunner._draft_load_format()` resolves it for a draft
runner and `build_load_config` takes it, so the `LoadConfig` is per-runner.
Model code also reads it off the bag while it builds — Inkling replaces
per-element noise in its shared-expert scales under dummy loading — so the
load is wrapped in a scoped bag override that puts the target's value back.
- `skip_tokenizer_init` was on the copy for nobody: `TpModelWorker` already
short-circuits the tokenizer for a draft worker (`or self.is_draft_worker`).
`PrefillCudaGraphRunner._max_addressable_prefix_len` capped the prefix by
`server_args.context_length`, which the copy used to carry for the draft; it now
reads the runner's own `model_config.context_len`. That is also more accurate for
the target, whose `--context-length` may be unset while the resolved context is
shorter than the token table.
What stays a variant is the dflash/dspark path's attention backend: backend
selection reads it off the config object the draft runner holds, and the
resolved gate has to survive the variant's publish. `draft_server_args_overrides`
now carries only those fields and says why.
`ServerArgs.override(source, **fields)` was the last way to change a resolved
`ServerArgs` in place. Every remaining call-site was one of two things, and
neither wanted an in-place write:
- **A config for someone else.** A draft worker's context length, an encode
worker's device, the compile script's watchdog, the client's port pick, a test
fixture's backends. These already deepcopied first — the write was on the copy.
- **A launcher-stage resolution.** `resolve_auto_parsers` detected the chat
template's parsers and wrote them back, to be inherited by the schedulers it
spawns.
Both are "one config becomes another", so `derive(source, **fields)` returns the
variant and leaves the receiver — and any bags projected from it — untouched. It
deliberately is not `dataclasses.replace`: resolution does not re-run, because
the values being set are decided after it, from inputs it never had. Provenance
and the resolvable-field stash work as before, on the copy.
`resolve_auto_parsers` now computes the parsers and returns the config to launch
with; the detection helpers stop taking a config to mutate. `HiMambaRadixCache`
re-applied a HiCache layout normalization `__post_init__` already performs (the
same duplicate removed from `UnifiedRadixCache` in ebb1c88d23) and just goes.
With no in-place mutation left, `ServerArgs.__setattr__` raising after
resolution *is* the guarantee, so the textual writer ratchet retires and
`test_server_args_derive.py` pins the contract instead: the receiver survives
deriving, the published instance still refuses assignment, and deriving does not
publish. `SGLANG_STRICT_CONFIG_MUTATION` was already unused — the guard has been
unconditional since the mutation sweep — and goes with it.
The detection tests drop their `SimpleNamespace` stand-in for a real
`ServerArgs`; the test kit and the MLA chunk-metadata fixture publish a derived
variant instead of writing the runner's published config.
`init_tokenizer_manager` wrote the chat-template-detected `reasoning_parser` /
`tool_call_parser` onto the published `ServerArgs`, after
`TokenizerManager.__init__` had already projected the config bags — so the
namespace readers and the resolved-config readback disagreed with the instance,
and a second `Engine` in the same process would inherit the first one's
detection through the shared bags.
Detection is per-engine control-plane state, which the manager already models:
`record_config_updates` records it, the readback endpoints overlay it, and
`config_value` reports what is in effect. `OpenAIServingChat` — the only reader
of these two fields in the tokenizer process — follows the overlay. The
architecture pass (`resolve_auto_parsers`, before the schedulers fork) is
unchanged: the scheduler resolves its own bags from the instance it receives.
`sa = get_server_args()` followed by `sa.field` reads the same startup record as
the direct form; the read ratchet added in the previous slice pinned twelve of
them as the remaining surface. Eleven now read the accessor for what they
actually want:
- `is_enable_moe_cp_allgather` compares the attention-CP and MoE-DP sizes to
decide whether a forward needs an allgather, so it reads the live topology
through `get_parallel()` — the same source `get_moe_cp_size()` right above it
already uses. Both groups exist once model-parallel init has run, which is
before any forward.
- The DeepSeek MLA decode-backend gate and Inkling's attention paths read
`get_exec().kernel`; Inkling's KV-dtype checks read `get_model()`. These are
per-runner fields, and the value they get is the config published for the
runner being built — unchanged from what the alias returned.
- The int8 mamba checkpoint pool reads `get_exec().mamba`. It keeps its guard
for callers that construct the pool with no published config; that guard now
catches the namespace accessor instead of the slot.
`model_loader`'s `moe_dp_size` stays on the instance and is exempt: the dict it
belongs to already reports the live size under `"dp"`, so that entry is the
configured intent, and `get_parallel()` shadows the name with the live value.
Alias-form baseline 12 -> 0. What remains on `get_server_args()` in the package
is the derived API (properties and methods computed from several fields plus the
HF config) and four config-intent reads of live-shadowed sizes, each exempt by
name with its reason.
`RayEngine.__init__` set the caller's `PlacementGroup` on `ServerArgs` as an
undeclared attribute, and every consumer read it back off the config object:
the non-DP launch path took `pg` and `is_custom_pg` from it, the DP path had to
re-attach the handle after `dataclasses.replace` dropped it, the DP controller
asked the config whether the group was the caller's, and the Ray HTTP launcher
had to clear the field so a stale handle could not leak in.
A live cluster object is not config. It now travels as a `placement_group`
argument on the two launch hooks — `Engine._launch_subprocesses` and
`Engine._launch_scheduler_processes` — defaulting to `None`, with `RayEngine`
holding the caller's group in `_placement_group` and the DP path passing
`is_custom_pg` down to the controller. The public API is unchanged:
`RayEngine(placement_group=pg, ...)` still works, `pg` still falls back to the
ambient group, and the HTTP launcher gets `None` from the default.
Writer ratchet 18 -> 15.
`get_server_args().<field>` reads one process's startup record. Nine sites still
did that for a value that has a namespace: the attention backend (5),
`skip_tokenizer_init` (2), the draft-aware `load_format`, and a chunked-prefill
size in `sglang.kernels`. They now read `get_exec().kernel` / `get_serving()` /
`get_model()` / `get_schedule()`, so they see the resolved value including
post-publish overrides.
The multimodal processor's device selection moves to the instance it was
constructed with rather than to a namespace: `base_gpu_id` differs per worker (the
encode-server DP workers each specialise their own copy), so no process-global
value can stand in for it, and engines sharing a tokenizer process each need their
own. Branch order, the NPU preprocess patches, and the case that leaves "device"
unset are unchanged.
What stays on `get_server_args()` is the derived API — `@property` and method
members computed from several fields plus the HF config
(`mamba_cache_chunk_size`, `get_model_config()`, `enable_mamba_extra_buffer*`) —
plus three config-intent reads of live-shadowed sizes, each of which needs an
answer the live topology property cannot give (the DSA indexer's PP gate must
short-circuit before touching the PP group, `allocation`'s DCP gate asks whether
DCP was configured at all, and the CUDA-IPC recycler runs where no group exists).
A new AST ratchet pins both shapes it can see — the direct call and an alias
bound from it in the same function — at 0 and 12 respectively, exempting the
derived APIs and those three sites by name. The alias-form baseline is not zero:
those reads are mostly per-runner fields in model code, and lowering them is the
next slice.
Two fixtures stopped faking config: `test_dllm_fdfo_kv_reuse` rebound
`allocation.get_server_args` to a SimpleNamespace, which silently stops
intercepting the moment a reader migrates; it publishes a real config instead.
When --forward-pass-metrics-ipc-name is left unset the reporter generates an
endpoint and has to hand it to external consumers (the documented contract is
that they read it back from the server config). That readback is the scheduler's
get_internal_state, which already reports get_context().resolved_server_args_dict(),
so the write moves to get_context().override and the read alongside it to
get_observability() — the endpoint still shows up in /server_info's
internal_states, and the ServerArgs instance stops being a message bus.
The test's server_args stand-in (a SimpleNamespace with a hand-rolled override)
becomes a real published config, so the reporter exercises the same accessors as
production.
Writer ratchet 19 -> 18.
The scheduler's runtime HiCache attach/detach wrote its own ServerArgs so the
internal-state readback would show the change; that readback already reports
the resolved config, so the writes become get_context().override(...) and the
namespace readers see them too.
The tokenizer side is per-engine — several Engines can share one process — so
its control-plane updates (weight version, model path + load format, HiCache
attach/detach) stay with the manager instead of moving to the process-global
bags. TokenizerManager gains record_config_updates / config_value /
resolved_config_dict, and the readbacks that used to observe the instance write
(/server_info, /model_info, the HiCache status endpoint, the gRPC bridge) now
overlay those updates onto the startup config.
test_server_info's stub grew the real manager instead of a SimpleNamespace, so
the overlay it now exercises cannot drift from production.
Writer ratchet 26 -> 19.
EAGLEWorkerV2, StandaloneWorkerV2, MultiLayerEagleWorkerV2 and
FrozenKVMTPWorkerV2 wrote the draft's context_length onto the ServerArgs
instance they share with the target worker, and the scheduler wrote the draft's
load_format onto that same object just before creating them. The target's config
carried draft values from then on, and anything constructed later in the process
inherited them.
Scheduler.maybe_init_draft_worker now makes one draft copy through
draft_server_args_copy() and hands it to both the worker factory and the worker,
so every algorithm gets it — the four built-ins, dflash/dspark (which deepcopy
it again inside build_draft_tp_worker), and anything registered through
SpeculativeAlgorithm.register. The copy starts from the config the process
resolved, not from the pristine seed, so load-time overrides made before this
point (the chunked-prefix gate, the SM100 GDN prefill default) are part of what
the draft sees; context_length and load_format are applied on top.
The construction runs under a preserved publish of that copy, the shape
build_draft_tp_worker already used. Weight loading reads the bags rather than the
instance it was handed — Inkling's ModelOpt scale normalization keys on
load_format — so the draft has to be built with its own config published, and
the target's is back in the slot when construction returns.
The EAGLE hot-token-map write is deleted, not moved. init_token_map runs from
alloc_memory_pool, long after the draft's TpModelWorker built its ModelConfig,
and hot_vocab_size is only ever read off model_config.hf_config, which
json_model_override_args reaches at ModelConfig construction. The write could not
affect the draft model; only the shared instance saw it. hot_token_id is
unchanged, so a draft checkpoint that declares hot_vocab_size behaves as before.
Tests: draft_server_args_copy carries the target context_length, a configured
draft load_format and any load-time override while leaving the target's instance
alone; and the scheduler handoff pins that the factory and the worker both
receive the copy, that the copy is the published config during construction, and
that the target's is restored afterwards.
Writer ratchet 31 -> 26.
Each of these wrote a value after resolution so a later reader would find it on
the instance. None of them needed the instance: one write was redundant, and the
two that carry a value the resolved-config readback reports move to
get_context().override, which the readback overlays.
- The SM100 GDN prefill default was written onto ServerArgs and read back one
line later by initialize_linear_attn_config. It is now the return value of
flashinfer_gdn_prefill_default, threaded into initialize_linear_attn_config
(an explicit --linear-attn-prefill-backend still wins) and recorded with
get_context().override so /server_info reports the backend in effect.
- The XGrammar fallback recorded grammar_backend="none" on the instance. No code
reads the field after the factory reads it once, but get_internal_state
reports the whole resolved config, so the fallback now lands there instead:
the readback tells the truth and the seed keeps the requested backend.
- UnifiedRadixCache.init_hicache re-applied the direct-IO layout fixup that
__post_init__ already applies: init_hicache only runs when hierarchical cache
is on, which is exactly when _handle_hicache normalizes page_first to
page_first_direct (pinned by test_hicache_io_backend_and_mem_layout_
compatibility::direct_with_page_first). Three fixtures reached the fixup by
building ServerArgs(model_path="dummy"), whose resolution is skipped, so they
now declare the layout resolution would have produced.
Writer ratchet 34 -> 31.
The skill still described the retired resolve-at-end contract (a single
resolved ServerArgs as the source of truth, ServerArgs.override as the
mutation entry). Rewrite the config sections for the landed model:
- ServerArgs is a pristine published seed; resolved config lives in the
namespace bags (get_exec()/get_memory()/.../get_device()), projected at
publish(server_args, role=...) per process entry.
- Post-publish mutation goes through get_context().override (bag-only);
ServerArgs.override is being retired behind the writer ratchet, and
rerouting a writer co-flips all its readers in the same commit.
- Documented the reads that stay on an instance (per-runner fork fields,
per-instance tokenizer/entrypoints boundaries, whole-object passes),
get_parallel()'s config/live dual semantics with the live-shadowed
sizes rule, preserve_config for nested publishes, and the bag-only
declare_load_time_override route.
- Testing idioms: publish-seeding (override_server_args + scoped bag
overrides) instead of faking accessors or SimpleNamespace stand-ins;
per-file test runs to avoid leaked-publish masking.
- Guardrails: added the writer ratchet, namespace-coverage lint, and the
migration-deferral ratchet.
publish(role=...) has recorded provenance since the namespace split; this
wires the enforcement the role was reserved for. SGLANG_ROLE_NAMESPACES
selects the mode:
- off (default): no bookkeeping; the mode gate in config_bag stays a
dead-branch-prunable check under dynamo (bag reads run inside compiled
forwards — pinned by a fullgraph test).
- record: audit mode — collect (role, namespace) pairs per process and
persist each new pair immediately to SGLANG_ROLE_NAMESPACES_OUT (worker
teardown skips atexit), plus a per-process stderr summary at exit.
- enforce: a bag read outside the role's ROLE_NAMESPACE_SETS entry fails
closed with an actionable error; None entries mean full tree.
Sets are filled only where audits back them: dp_controller reads only
exec (record-mode plain + DP-attention smokes agree with the module's
static read set — the elastic-EP gate). tokenizer observed zero bag
reads (per-instance managers read self.server_args by design) but keeps
the full tree until the multi-tokenizer disagg shape (TokenizerWorker's
get_disagg read) is audited; encoder / expert_backup /
weight_cache_daemon likewise await their deployment shapes.
Verified end-to-end: DP-attention smoke under enforce boots and serves
with zero violations.
The module-skipped tests injected config by faking get_server_args (a
SimpleNamespace stand-in patched onto the module) or by writing fields
onto a ServerArgs instance post-publish — both invisible to the namespace
accessors the production code now reads. Re-enable them by publishing the
config they need (get_context().override_server_args seeding, scoped per
test), asserting bag state where the old assertions checked instance
write-through (declare_load_time_override is bag-only), and extending the
per-runner stubs the code genuinely reads (kv_cache_dtype_str,
max_total_tokens, context_len).
The unified-radix-cache file (which grew a large hicache/insert-walk suite
while skipped) is recovered in the same change:
- test_cache_finished_req_strips_thinking (19 parametrized classes) wrote
strip_thinking_cache onto the ServerArgs instance; the cache reads
get_serving().strip_thinking_cache — use the serving bag's scoped
override.
- test_shallower_crossing_backs_up_above_backuped_middle staged its
broken-backup-continuity setup through insert_host, which now
deliberately drops refills below an un-backed-up node under
write-through (host_insert_dropped). Build the same tree state through
an explicit backup + device eviction.
Every config-namespace-migration deferral is recovered, so the deferral
ratchet (test_migration_deferral_ratchet.py) has done its job and is
retired.
The parallel namespace joins the accessor migration: 106 config-leaf reads
(enable_dp_lm_head, enable_dp_attention, pp_async_batch_depth, dp_size,
ep_join_rank_offset, dwdp_size, ...) flip from get_server_args()/
self.server_args to get_parallel(), which serves config leaves from the
published parallel bag via __getattr__.
- ParallelContext.__getattr__ is restructured to stay dynamo-traceable
(object.__getattribute__ graph-breaks): gate helpers such as
enable_moe_dense_fully_dp() run inside compiled model forwards. A
fullgraph regression test pins the pattern.
- The five live-shadowed topology sizes (tp/pp/dcp/attn_cp/moe_dp_size)
keep their server_args reads: the live @property wins on the accessor,
and conditionally-initialized groups would fail loud at unconditional
call sites.
- Elastic-EP scale writers (ep_size/dp_size x4 in model_runner) reroute
to get_context().override together with their remaining instance
readers (expert_location gpus-per-node paths); the ServerArgs.override
ratchet drops 39 -> 35.
- The expert placement helpers (compute_logical_to_rank_dispatch_
physical_map, _compute_logical_to_all_physical_map,
_prefer_same_node_experts) now read everything from the bags and drop
their server_args parameter; their unit tests publish the config they
need instead of stubbing it.
The load-time gate (maybe_disable_chunked_prefix_cache) wrote its
ServerArgs instance while every reader has moved to the published
config: the attention backends assert / branch on
get_schedule().disable_chunked_prefix_cache when they initialize, so the
flip never reached them and a backend outside
CHUNKED_PREFIX_CACHE_SUPPORTED_ATTENTION_BACKENDS kept chunked prefix
enabled.
Reroute the writer through get_context().override (which writes the
published bags) and flip the two remaining instance reads — the gate's
own log check and the prefill cuda-graph runner's capture flag — to the
bag. A regression test pins the three contracts: the gate lands on the
bag, the pristine ServerArgs instance stays untouched, and the
draft-worker guard never writes.
The ServerArgs.override call-site ratchet drops 39 -> 38.
