Four ww tests carry the last eight native byteid carriers' assertions:
asmwindow (753 convwrap beta/main/alpha order + window polarity, 754
slice stride + negative scale scan, 755 amp-dot-idx four rows, 758
first-CALL-line extraction + tab-framed disp literals, direct
w6c/w6c_ww never ww build), freenoop (930 byte-id strengthened to
per-stream compare + negative 'free' grep), structabi (946 param/ret
MOVSD windows with polarity tables and SEND/RECV agreement), mangle
(989_m1mangle needles + concat byte-id, glob order strengthened to
byte-lexicographic). Dead want/stage_mask row fields documented, not
invented into runtime legs; the w6c_ww-absent skip gates drop because
the Make target declares the tools.
BYTEID_WRAPPER_SOURCES, its bins, and test-native-byteid are deleted —
the native byte/artifact partition is EMPTY (11 -> 0 this session);
docs counts move to 123 carriers.
25 renames (git mv, content untouched). _test.ww is what the package
coordinator's test detection and the sep loader's canonical exclusion
key on; the old *test.ww spellings survived only through the
line-leading-@test compatibility scan. Consumers updated in place:
LIBRARY_TESTS, the libbyteid roster, the 901/974/975/976 carriers that
copy or invoke these files, and the check.c/check.ww + path/ftos
comments that cite them. Closes the open-driver-work migration bullet.
test/byteid/wwi_test.ww carries every assertion of both carriers: the
M2 producer gate (ascii/strings/getopt driver-combined positives with
byte-id + wwdump -a re-parse, the synth decl-kind/type-node sweep with
the #47 @symbol round-trip, the #48 types export-def gate, and the
negative private-nominal leak with both-stage identical reject
diagnostics), plus the #11 wwileaf decl-less-leaf rows (package-line
leaf, cross-stage .wwi byte-id, importer resolution + .s byte-id).
Byteid carriers 10 -> 8; docs counts move; the byteid ww target gains
the wwdump prereq its re-parse legs invoke.
test/testenv (package testenv) is the shared plumbing for ww-native
test drivers — subprocess launch with captured output, file IO, string
search, byte-sorted directory listing, scratch ownership — lifted from
the proven package_test.ww idiom; a utility, not a framework.
test/byteid/libbyteid_test.ww carries every assertion of the retired C
carrier: the 44-entry roster (fixtures + sentinel-guarded import probes
+ the zero-dep root-only build), both-stage sep builds with the
resolved-unit proof, per-package .s concat compare with the empty-concat
guard, the ID/DIVERGE/WWREJECT pin discipline, and the lib/ corpus
completeness scan (negative-verified against a planted un-enrolled
module). Concat order strengthened from shell-glob to explicit
byte-lexicographic. The roster fills by cursor-driven field writes
because append() rejects struct-call-result sources (#34).
Byteid carriers 11 -> 10; docs counts move; test-byteid runs the ww
test via the wwtest/ pattern with WW_TEST_REPO.
One phony target per test file so -jN parallelizes both suites, each
file building in its own persistent workdir so only changed packages
recompile. Tests always execute — the workdir caches builds, never
results. Lang tests keep their exact historical invocation (no -I);
library tests keep -I lib/ww. Warm -j4: test-library 19.0s -> 1.0s,
test-lang 19.8s -> 2.0s.
Cold populate, full-skip rerun, surgical staleness (one package's
artifacts deleted), and a 0-byte .o poison drill each build the w6a
graph through a persistent workdir and must be byte-identical to a
classic fresh-scratch build, on both driver stages, plus the
cross-stage clean compare. Wired into test-all; the inner-loop targets
stay untaxed.
Per-binary source lists corrected against the real import closures
(strace of each cold sep build): every wwstage tool gains
lib/types/types.ww, wwdump_ww/w6c_ww gain lib/math/{floats,math}.ww and
drop the no-longer-imported lib/fmt/fmt.ww, ww_ww drops
lib/strconv/strconv.ww, and WWTEST_SRC gains the strconv split plus
math. Wrapper test binaries link only libwcc.a — the runtime tools they
exec move to the run targets that launch them, so a .ww edit no longer
relinks 111 carrier binaries. test-compiler-smoke declares the
w6a_ww/w6l_ww/ww_ww fleet its WW cell actually drives (latent
fresh-clone gap); test-package replaces the blanket all with its exact
tool set; test-lang/test-library name the cstage subset they invoke.
The five wwstage tools, wwtest, wwfixture, both exec-test binaries, and
the package-test binary move off the per-invocation mktemp workspace to
one persistent ww-build workdir per target under out/wwbuild/, publishing
into out/bin by one atomic mv. Package artifacts now survive across
rebuilds and reuse on exact content identity, so a toolchain-source
touch rebuilds in seconds instead of five from-scratch sep builds; the
per-target split keeps -jN recipe concurrency race-free. Rebuilt
binaries verified byte-identical to the mktemp-path builds for all ten
targets.
A -w DIR workdir replaces the fresh .sepwork scratch with a caller-owned
persistent package-artifact tree. A package is reused only when its
freshly composed unit byte-equals the committed unit and byte copies of
the compiler/assembler recorded in the dir equal the live tools — pure
content identity, no mtimes, no hashes, every decision reproducible
with cmp against plain files. Recompiles stage at .new names and commit
by rename, unit strictly last, so an interrupted build forces a
recompile and can never leave a committed unit vouching for uncommitted
artifacts; .o/.a additionally reject zero size (ELF/ar are never
empty), while .s/.wwi accept legitimate empties (FFI-only rt). A mode
stamp pins the -T/-S shape and the artifact protocol revision. Classic
scratch keeps its exact acquire/refuse/cleanup contract; run rejects
-w; dir-mode test rejects -w; both driver stages implement identical
behavior and wording.
Ten of the thirteen were already fully twinned — their runtime rows
migrated in earlier waves (r56_def_*, r79_*, r78_*, r926_*,
r989_m1union_*) and the carriers survived only for a cstage-vs-wwstage
byte-compare the blanket test-data-byteid comparator owns once the
source is a corpus fixture. The other three needed their remaining
inline sources added first: r756_* (4, alias-chain unwrap),
r844_bid_src (size(u64) untyped-int compare), r989_structframe_* (3,
struct-local frame layout). Every new fixture was verified both-stage:
compile exit 0 twice, .s byte-identical, runtime exit as pinned.
The five ww_ww-driver-leg carriers among them (785/786/789/790/
m1union) follow the 815/940/951 precedent: content identity owned by
their corpus twins, driver-leg identity owned by 989_lib_byteid's
sweep. Byte/artifact partition 24 -> 11; carriers 147 -> 134; corpus
pin 1495/2990; the stale 915 sweep figure in the docs corrected to
the true 1,157 non-error count.
fnptrcalleetfn's N_DOT arm required an N_IDENT base, so a.b.cb(...)
fell to the name registry with an empty module hint and emitted
CALL cb(SB) (undefined symbol; cstage calls the stamped ptr indirect).
dotbasestructinfo resolves the base chain through the struct registry
— each link a struct- or *struct-typed field — and the single-dot
path routes through the same resolver unchanged. Closes the last open
shape of the #59.8 name-keyed callee family. Fixture fnptrfield_chain;
corpus pin 1487/2974.
installparams normalizes a decl's `T...` param lhs to []T in place
(cstage instead sets the resolved tp->type and never mutates AST), so
typeeqast compared a decl-synthesized fn type's []i64 against a fn
TYPE expr's surface i64 and confidently rejected
`let f: fn(args: i64...) void = sum;` — cstage accepts and runs. The
normalization wrapper now carries an op marker and typeeqast peels
exactly it, so both sides compare at the declared element type without
admitting genuinely different element depths. Fixture fnptrlet_variadic;
corpus pin 1486/2972.
The last frontend-gap pin: wwstage had no Hare struct embedding
(struct { hash.hash, ... }), rejecting lib/crypto/sha256 at parse.
- parse.ww: the three member forms (named / anonymous struct / bare
dotted-ident embed), consume-then-branch since this parser has no
peek; embeds carry f.str == "" and the type in f.lhs.
- check.ww N_TSTRUCT flatten: promote the inner struct's flattened
fields at base+src.offset (check.c:961-990); the embed is one
nested-struct unit in the slot ladder; the resolved inner AST is
planted on the TFIELD rhs for cgen.
- check.ww walkers: astoffset / exprtype N_DOT / #251 struct-lit
field lookups descend embeds through shared helpers; the collision
and non-struct-embed rejects live in validatestructfields (the
once-per-decl diagnostic site).
- cgenutil.ww registerstruct: regfieldrun walks the AST against the
flattened tfield cursor, descending embeds via the planted inner
AST so promoted fieldinfo entries keep the inner field's own name
and type node.
- wwi printers unchanged (both stages already emit nameless fields).
sha256_test compiles byte-identically end to end and its 6 tests
pass; 989_lib_byteid is now 44 id / 0 divergent / 0 wwreject.
Fixtures r5913_* (promoted rw, offset shift, anonymous embed,
two-level embed + promoted fn-ptr callee, three rejects); corpus pin
1485/2970.
fnptrcalleetfn's N_DOT arm accepted only a bare N_TFN field tnode, so
a call through a field declared `cb: *fn(...)` missed, fell through to
the name registry, and emitted CALL cb(SB) — an undefined symbol
(loud at link; cstage calls the stamped ptr-to-fn indirectly). Same
TPTR peel the N_IDENT arm already had. Fixture fnptrfield_call covers
the by-value and via-pointer shapes; corpus pin 1478/2956.
The migrate-and-retire audit left the surviving carriers as the only
files still describing the retired combined.ww amalgamator as live:
m2wwi/m3sep headers claimed .wwi was dead code, four byteid carriers
carried dead cleanup of never-written .combined.ww intermediates, and
sepbuild's enumeration mirror plus lib_byteid's completeness scan
still skipped the retired artifact form (the same silent
accommodation just removed from both drivers — dropping it here
surfaced and flushed 39 stale untracked amalgamator outputs across
lib/, which the gates now reject loudly). Remaining mentions are
past-tense history or quoted diagnostics.
Audit-driven migrate-and-retire wave (one read-only auditor per
carrier batch, verdicts row-checked through the full corpus gate):
every retired carrier's assertions are now owned by declarative
fixtures or an existing owner. 238 new fixtures land (corpus 1,239 ->
1,477; cells 2,954), covering enum/str/slice/tagged array elements,
inferred-length arrays, global slice/str zero-init and literals,
global array/pointer field reads and stores, struct-literal slice
fields, struct returns, float arithmetic, size/int limits, sort and
log-vstream behavior, place-addressed stores, and append places.
Rejection rows STRENGTHEN the old nonzero-exit checks to required
diagnostic fragments; all byte-id claims fold into the blanket
test-data-byteid sweep (now 1,142 compared, 0 pinned-divergent).
Four carriers were fully redundant with existing fixtures/lang tests
(749, 722, 765, and 771's byteid rows) and retire without new rows.
Native carriers 171 -> 147 (residual 135 -> 111).
PROJECT_PLAN.md was a historical disclaimer wrapping five still-open
T1 driver items — those move to docs/test-system-v2.md (Open driver
work); the rest is dated 2026-08-05 design history, archived in git.
Also retired-mechanism cleanup: both drivers drop the *.combined.ww
enumeration skip (the amalgamator is gone; leftover debris now fails
loudly instead of being silently accommodated), the consumer-less
m4_combined_epoch.md5 pin is deleted, and the wwi.ww/ww.h/check.ww
comments that still described combined.ww as the live path are
re-cited to the sep reality.
The wwstage variadic call classification was name-keyed: a fn-ptr
FIELD call whose local base shadowed the current module name
(lib/log's log.println(log, args...)) picked the module fn's
signature — nfixed off by one, the fixed arg boxed into the gather,
the spread emitted as zeros — and a no-collision fn-ptr callee missed
the registry entirely, leaking the raw N_SPREAD as a single $0 word
(SIGSEGV / exit 255 in 8 of 11 logtest tests on the wwstage leg).
fnptrcalleetfn resolves a local fn-ptr callee (bare local or struct
field) to its N_TFN once, shared by the CALL-target choice,
callee_variadic_param (with the []T wrap registry params get from
installparams), calleecvariadic, and the widening param lookup, so
target and arg prep can never disagree. Graduates the #59.8 logtest
pin — DATABYTEID_DIVERGED-era M_DIVERGE count is now zero.
A TYPE name used as a VALUE (an error-singleton `return too_long;`)
stamped the flattened BODY type, so structurally identical !void
singletons in one union were indistinguishable and flatvariantidxt
loud-rejected the ambiguity — the real cause of the lib/path wwstage
reject (the pinned #120/#29 global-slice-const blame was stale; that
family had drained). The N_IDENT arm resolves through a synthesized
TNAME (the #66 N_STRUCTLIT precedent); the module-qualified N_DOT twin
reads the sym's cached NAMED. Graduates the path M_WWREJECT pin (#142).
An enum OR-fold under `as` ((m.A|m.B) as u32) was never typed:
scruttype resolves IDENT/DOT only, and the post-order restamp then
re-derived the folded member literals as untyped_int. The unstamped
operand missed cgtypeassert's #27b enum-reinterpret gate and lowered
as a phantom tagged assert — unconditional exit 1 at runtime on the
wwstage leg. checkisas now types an N_BIN lhs and the N_BIN restamp
preserves an existing enum stamp. Graduates the #59.9 stattest pin.
The wwstage nodeisstr/nodeisslice recognizers were name-keyed for
every non-local shape: an indirect fn-pointer callee ((*f)() — the
errnotest #59.5 divergence, streq receiving a shifted register file)
and module-global let/const idents (path.sepstr — union-widen pushes
zero-filled len/cap) both fell to false while cstage keys off the
checker stamp unconditionally. Both recognizers now fall back to the
stamped n.type_; push and pop sites share them, so the drain stays
balanced by construction. Graduates the #59.5 errnotest pin.
test/package/runtime/ was a manual README-driven corpus with no
target. Its unique surface — the driver's single-file ww test leg,
the unmet-expected-abort outcome, and the published -c binary's own
CLI (repeated globs, -list, -timeout-ms, direct-binary timeout and
descendant handling) — is now asserted by three new @test fns in
package_test.ww; the README drops its manual-run instructions.
Wwstage-driver-leg byte identity gets one owner. The three carriers'
content identity is already blanket-owned by their r815_/r940_/r951_
corpus twins via test-data-byteid; their only non-redundant assertion
was the ww_ww driver leg, which 989_lib_byteid's 43-unit sweep covers
apart from the zero-dep root-only edge — folded in as a new probe row
(44 units). r940 fixture comments re-cited off the retired carrier.
x.slicefield[:] / x.strfield[2:] emitted MOVQ $0 for the omitted hi
bound on BOTH stages (N_IDENT-gated dispatch; symmetric, so byte
identity never caught it) in all four sites: cgexpr N_SLICE + pushargs
(cstage), cgslice + pushargsrev (wwstage). The new arm re-evaluates
the pure field read for its {ptr,len,cap} header and takes .len,
covering local, viaptr, dot-chain, (*p), arr[i], and global inners.
Call inners still loud-reject upstream. Sibling of the #252/#257
array-field arms.
let x: []T = g.buf on a module-global struct dispatched only str
fields to the 3-word header load; a []T field fell to the scalar
tail (ptr word only), so len/cap read stale registers and the shape
was byteid-divergent against cstage's TY_STR||TY_SLICE arm (#263).
The four graduated pin families byte-compared under test-data-byteid
but their original fixtures masked the miscompiles at runtime (lone
frame slot, dead code path, pointer accident). Add the discriminating
programs as run fixtures: adjacent-element clobber for the 2-byte
deref store, second-variant match dispatch for fn-type dedup, payload
integrity across the tagged widen of a padded struct, and base64
clear() actually zeroing its buffers. Corpus pins move to 1,229
fixtures / 137 run / 766 run-exit / 2,458 cells with the new identity
hash; the data-byteid floor moves to 915.
The N_TFN arm stamped only size and return, so every fn tinfo carried
a nil param chain and typeeq judged any two same-return fn types
equal: tagged-union dedup collapsed (*fn(A) T | *fn(B) T) to a bare
8-byte pointer and match read the pointer word as a tag, falling
through every arm for a real second-variant value. Build the tparam
chain like cstage's N_TFN resolve (bare ... sets the FFI variadic
flag; a Hare T... param wraps to []T with a per-param variadic bit
that typeeq now compares, mirroring cstage type_eq). Graduates the
four r76_typeeq_fn pins; the DATABYTEID_DIVERGED ledger is empty.
cstage types the binding ty_u8; the wwstage single-binder arm peeled
the element only for slice/array iterables, so a str scrutinee fell to
the N_FORRANGE fallback decl and exprtype stamped the binding str.
The stamp-keyed call-arg marshal then pushed the 3-word str ABI for a
1-word scalar, and an uncast compare on the binding was falsely
rejected. Route str through the same synthetic-N_LET binder with a u8
element. Graduates r940_str_forrange_arg.
cgwidentaggedstorebp's struct-ident arms (local and module-global)
counted copy bytes from structinfo.totsize, the round-8 slot width, so
a struct whose natural size is not a multiple of 8 took the MOVQ word
loop and copied the source slot's dirty pad bytes over the zero-fill
the arm itself just established; the width-true tail emitter beneath
was dead code. cstage keys the same copy on su->size. Route both arms
through copysrcnatsize (the stamped-tinfo natural size, #71).
Graduates r71_tagged_return_scratch_mixed.
pushargsrev's N_SLICE arm resolved esz and the base address from the
stamped tinfo of an N_DOT [N]T-field base but let the omitted hi bound
fall to $0, so the #258 array-to-slice borrow desugar handed callees a
zero-length slice: base64 clear()'s bytes.zero(e.ibuf) was a runtime
no-op under the wwstage frontend. Take the element count from the
field's array tinfo like the cgslice #252 arm and cstage's bu->alen.
Graduates r989_libprecond_decodedsize_aligned and the 989_lib_byteid
#59.3 base64 pin.
The plain `*p = v` arm's storeop map handled 1- and 4-byte pointees
but let a 2-byte one fall to MOVQ, an 8-byte over-store that clobbers
six adjacent bytes (masked in the pinned fixture by the lone frame
slot; loud with `&a[0]` of a [4]i16). cstage fldstoreop maps 2 to
MOVW. Graduates r660_i16_out_param_negative.
Every probe's assertion is owned by a current gate: the compile and
link probes by make all and the bootstrap fixed point; build/run and
cs/ww byte identity by the fixture corpus, test-data-byteid, and
989_lib_byteid; wwstage driver and toolchain parity by 993/995;
checker-diagnostic parity by the corpus' both-stage //ww:error rows.
The wwdump -t/-a dump-parity probes gated the frontend port's
convergence, which the compiler-output identity gates now own end to
end; carrier ran green at retirement.
What was still uniquely alive migrates: smoke.ww becomes corpus
fixture selfhost_smoke (upgraded from a cstage-only build to both
frontends, byte-identical, exit 42 on both toolchains; corpus pins
move to 1,225/763/2,450 with the new identity hash in the same
commit), and sym_link.ww's scope/sym behavior rows become in-language
lib/ww/syntax/symtest.ww under LIBRARY_TESTS. uses.ww (parser-stub-era
-a fixture) and the already-orphaned tagged_ptr_ret.ww/trypromote.ww
retire with the probe corpus. Bootstrap native gates drop to six;
frontend numeric-sync comments now cite the rule-6 mirror instead of
the retired diff probe.
Same contract as the 27 repaired byte/artifact survivors: checked
acquisition, one all-exit cleanup funnel per carrier, ENOENT-tolerant
checked unlinks, exact-path deletion, and cleanup failure fails a
passing carrier without overwriting its diagnostic. Also closes the
vacuous-green channels: empty-artifact pairs no longer byte-compare
equal, capture and staging failures fail their row loudly, 950 requires
a clean checker exit on no-error rows and hard-fails on a missing
wwdump_ww, 992 keys its scratch per invocation so a failed cleanup
cannot leak stale objects into the next tool's compare, and 995 no
longer pre-sweeps a workdir it never created. No assertion is weakened;
990_selfhost is retired in the next commit rather than repaired.
test-system-v2.md is the authoritative owner table and target graph;
CLAUDE.md rule 13 moves from the sizelint gate to authoritative layout
helpers and rule 14 names the new targets; PROJECT_PLAN.md is marked
historical where superseded.
peel-ok/sizelint-ok/primsize-ok annotations lose their tools; sites
keep the WHY in plain words. Citations of retired carriers move to
their fixture or @test successors (949_errtype_compare -> r949_*,
900_stdlib -> library owners).
Presence-is-registration wrapper classification replaces the 296-entry
TESTS list, its per-wrapper rules, the test/run scheduler, the
last-green cache, and auto -j; test is five in-process units plus one
fixture smoke; test-commit composes behavior suites; byte identity,
bootstrap, and platform stay explicit gates. test-data-byteid builds
every non-error corpus fixture twice through the fixed cstage driver
swapping only WW_W6C and byte-compares every per-package .s, with a
910-fixture vacuity floor and the DATABYTEID_DIVERGED loud-pin ledger
(8 real cs/ww divergences: r660, r71, r76_typeeq_fn x4, r940, r989 -
the base64 #59 pin). The 61 carriers whose only assertion that
comparator subsumes retire with it, leaving the 27 survivors that
observe asm patterns, symbols, frames, .wwi round-trips, inline-only
sources, or the wwstage driver leg.
One commit because the wildcard classification sweeps any leftover .c
into test-compiler, where the retired carriers' --sep invocations
cannot run against the current driver; the 61 deletions, the 27-entry
byteid list, and the blanket comparator are a single consistent state.
tools/peellint and tools/sizelint (and the 944 gate wired to them)
retire with the target graph that carried them; rule-13 layout
discipline stays on the authoritative-helper rule.
Behavior and reject carriers whose rows now live as test/wcc/data
fixtures or test/lang @test tables; the 31 library launchers plus
900_stdlib, whose only assertion was an existing @test source's exit
status (LIBRARY_TESTS runs those sources directly); and the test/run
shell harness, its protocol self-test, and the runww corpus runner.
Their Makefile registrations dangle until the target-graph flip in
the next commits.
local_zeroinit_test.ww takes the 840/944 zero-init seam (dirty-frame
prime and probe share one @test because the runtime forks per test);
alias_cgen_b6, alloc_nested_field, array_static_init, and
strarray_static gain the rows their retired wrappers held; the
remaining files re-point reject-row citations at the r788_* and
stage-matrix fixtures.
Every surviving carrier now owns its artifacts: checked mkdir/mkdtemp/
fopen acquisition, one all-exit cleanup funnel per carrier, ENOENT-
tolerant checked unlinks, exact-path deletion (rm -rf only for an
owned pid-keyed dir or a .sepwork beneath one), and cleanup failure
fails a passing carrier without overwriting its diagnostic. In the
same pass the carriers adapt to the driver contract this branch lands:
--sep and WW_PKGCACHE are gone, -S and the /tmp/ww_run_<pid> scratch
contract are asserted, and rows whose runtime or reject coverage moved
to test/wcc/data fixtures or test/lang @test owners are trimmed to the
byte/artifact/diagnostic observations only they can make.
Repair and adaptation ride together because most files interleave both
in the same hunks; splitting would manufacture intermediate carrier
states that never existed and cannot run against either driver.
000_smoke drops its popen ww -V leg (driver version parity moves to
949_driver_flagargs); 400_w6c feeds its sources through lex/parse/
check/cgen into open_memstream instead of spawning w6c. The five unit
binaries now hold zero subprocess calls.
Discovers test/wcc/data, hard-pins corpus identity (1,224 fixtures,
314 error / 12 compile / 136 run / 762 run-exit, sha256 name hash),
schedules both frontend cells per fixture through os.exec with
per-cell capture dirs, and emits the strict wwfix TSV result stream.
integration.sh is the black-box owner of the CLI/protocol boundary.
The r-prefixed waves absorb the runtime, reject, and byte-compare rows
of the migrated native carriers; each fixture is one directory with one
case.ww and a //ww:error, //ww:compile, //ww:run, or //ww:run-exit
directive covering both frontends.
Scratch acquisition is one bare mkdir that refuses a pre-existing
.sepwork; the path reaches cleanup only after acquisition succeeds.
ww build and ww test -o keep the tree for the caller; ww run and
no-output ww test build under an owned /tmp/ww_{run,test}_<pid> dir
and remove it on every result, escalating rc on cleanup failure.
The content-keyed package cache is deleted from both stages; -S stops
after per-package .s emission; --sep is gone and unknown flags are
rejected; directory ww test delegates to the wwtest coordinator; the
wwstage driver routes every subprocess through os.exec.
The wwstage Makefile recipes move off the persistent <tool>_ww.d
build-in-place shape to a per-invocation mktemp workspace: the old
recipes relied on the retired pre-wipe semantics and any rebuild
would trip the new refusal on their leftover main.sepwork.
The three 989_pkgcache_* carriers lose their subject with the cache
and go with it; the rest of the wave's carrier adaptation follows in
its own commits.
