t27o: a fn returning Vec<T> built by push converts in the out-buffer form -- a last out: []T, the result the count, the oracle checks the written prefix (Closes #11073, Closes #11112) by gHashTag · Pull Request #11109 · gHashTag/t27

gHashTag · 2026-10-19T17:22:41Z

Step 2 of #10546 (part of #10522; step 1 is #10648). A fn that returns Vec<T> and builds it by push / extend_from_slice converts in the out-buffer form, the move #9098 made for an owned String. The last commit fixes #11112, which this form exposed (see "#11112" below).

The rule, and where it lives

specs/tools/t27o/rust_convert.t27 (header paragraph "The Vec out-buffer form", the section "the Vec out-buffer form (#11073)", vectors / put_vec_record for the oracle). gen/rust/tools/t27o/rust_convert.rs is t27c gen-rust of it; no hand-written Rust, no cli/ change.

A fn whose result is Vec<T>, T an integer, float or bool type, whose body makes the Vec as one local (Vec::new(), Vec::with_capacity(n), vec![]), changes it only by push(e) and extend_from_slice(s) and hands it back (return v, the tail v, an empty Vec in an early return), takes a last out: []T and returns usize, the number of elements the Rust fn returns:

fn evens(n: u32) -> Vec<u32> { let mut v = Vec::new(); .. v.push(i * 2); .. v }
fn evens(n: u32, out: []u32) -> usize {
    var t27o_w : usize = 0;
    ..
    const t27o_e104 = i * 2;
    if (t27o_w < out.len) { out[t27o_w] = t27o_e104; }
    t27o_w = t27o_w + 1;
    ..
    return t27o_w;
}

The capacity is the caller's and is never guessed: an element past out.len is counted and not written (the String form's convention), so the count is the same whatever room the buffer has. with_capacity(n) is a hint: read for what reading it does, then dropped.
The element of a push is evaluated once, whether it fits or not (as Rust does).
Refused, each with a reason that names the use (new R_VECOUT): a read of the Vec (len, an index), any other change (insert, pop, sort, an index assignment), a borrow, a rebinding, a second Vec local, a result made another way (a vec! literal, a clone). A Vec that grows without a bound the caller states has no t27 form; the reason says so.
A call of a fn in this form is refused (new R_VECCALL): the caller would own a buffer of no stated length. (The String form leaves its callers to call it without the buffer, see "Found, not fixed".)
The oracle records the count and every element of the written prefix (a result of at most 256 elements) in a 256-element buffer, and asserts that the same call with a zero-length buffer answers the same count.

Measure (the five repositories at the commits of step 1: t27 `7496ed2`, trios 483dd9ee, trinity 0770abb9, trinity-fpga 20bf4c418, 999-multibots-telegraf 6e5512881; 1,388 tracked `.rs` files, `gen/` and `target/` left out; the converter built from the generated Rust, one file per process, 60 s cap)

How the 626 fns that return a bare Vec<T> (360 outside #[cfg(test)] modules; 123 more return an Option / Result of a Vec) make the result:

result made by	fns
a `vec![..]` literal as the tail	251
`collect()`	135 + 16 (a local started by one)
push / extend only, from an empty local	91
a local started as a vec! literal, or by a call	44 + 4
a delegated call	25
other Vec methods (`sort`, `insert`, ..)	23
push plus reads of the Vec	11 + 1
a clone / copy, an empty literal, other tails	10 + 3 + 12

Of the 91 push / extend fns the element type is an integer, float or bool in 19, a struct or enum of the file in 32, a tuple in 21, a text in 15, other in 4. This form takes the 19 (the oracle can record only scalars).

first refusal of a fn (the five files `bootstrap/src/compiler.rs`, `service.rs`, `cli/t27b/src/lower.rs`, `cli/tri/src/fpga.rs`, `gates.rs` left out, as in step 1)	before	after
type `Vec`	1,126	1,108
a Vec result that is not one local built only by push (R_VECOUT)	0	4
a call of a fn whose Vec result is the caller's out buffer (R_VECCALL)	0	1
macro `vec`, a vec! the fn changes / hands on, element type or length unknown, path `Vec::new`	529, 75, 19, 71, 66	unchanged
uses refused (cascade)	3,717	3,718
fns converted of 16,765	1,653	1,657

All files: converted 1,820 to 1,824 of 19,392. The 18 fns that left type Vec (outside the five files): 4 convert (from_bits, to_bytes, uniform_lengths, regex_lite_global), 3 are R_VECOUT (they read the Vec: crc32(&buf) in both encode, iter in self_attention), and 11 pass the Vec and stop at the next refusal. Over all files the first refusal of 25 fns moved: 4 convert, 4 are R_VECOUT, 1 is R_VECCALL, and 16 stop at the next refusal (to_le_bytes / to_be_bytes 5, lines 4, a refused self 2, five others). So this form reaches 4 fns of 626 today, and the numbers say why: the Vec-returning fns are mostly literals and collects, not pushes.

Measured and not shipped: allowing struct, enum and &str elements converts 8 more fns, but the oracle cannot record them element by element yet, so the form stays at scalars.

Per-file outputs: 1,368 of 1,388 are byte-identical to the merge base; the 20 that differ are exactly the files of the fns above and their callers. Dogfood, as in #11064:

Measure	Before	After
the 29 files of the night's sample list: fns seen / converted	571 / 40	571 / 40 (1 output differs: `firmware.rs`, a refusal reason)
every tracked `.rs` of t27 at `7496ed2` (481 files, `gen/` included; `compiler.rs` timed out at 60 s under load in both runs): fns seen / converted	10,522 / 3,052	10,522 / 3,053 (9 outputs differ)
`.rs` of trios, trinity, trinity-fpga, 999-multibots-telegraf (1,006 files, `gen/` left out)	9,657 / 1,025	9,657 / 1,028 (10 outputs differ)

Replay of `--check` (rustc harness, run, `t27b test` of spec plus samples), measured on `46b3eff`, the head before the rebase (the t27o files did not change on master since; the rebased `gen` file is byte-identical to `t27c gen-rust` of the spec)

Oracle harnesses: 9 of 1,388 differ. Four because of a Vec result (bitstream.rs +5 sample calls, padding_length_distribution_guard.rs +1, trios-phd/src/main.rs +8, jepa_t_ingest/src/lib.rs the converted item text only). Four because of the comma fix below (5 zero-parameter String fns). compiler.rs differed only because the 60 s cap hit under load; alone, its harness is byte-identical.
Replayed end to end, base and head converters, the 17 files whose conversion changed and that fit: 14 have the same verdict (jepa_t_ingest fails the same oracle_token_to_q15 u16 as i16 before and after), bitstream.rs PASS with 0 tests to PASS with 1, trios-phd/src/main.rs PASS with 3 to PASS with 4, and one file went the wrong way at that head: padding_length_distribution_guard.rs PASS with 0 tests (vacuous) to t27b "unsupported" and t27c test-report BLOCKED (tuple field index must be comptime-known). That was the untyped array constant of [bug] t27o: a read-only integer array local iterated by reference converts to an untyped array that t27c rejects #11112, below; after the last commit the file is a PASS with 1 test. Files that go from pass to not pass: 0.

Found by the replay, fixed here (each test written first and failing)

The first harness of regex_lite_global (s[i..].find(needle) with the sampled needle "" pushes for ever) ate the memory of the machine. A Vec result is an input-dependent allocation, so the harness now carries t27o_guard when some fn is in this form: the original runs on a thread with 3 s and a counting global allocator that parks the thread at 16 MiB; the sample is skipped like one on which Rust panics, two such samples end the sampling of that fn. Verified on a small program: a runaway is skipped in 3 s and costs 16 MiB.
uniform_lengths() was written uniform_lengths(, t27o_o0[0 .. 256]): the comma before the buffer was written even with no argument. The String form had the same latent defect (chrono_now_iso(, t27o_o0[..]); 5 fns in 4 files of the five repositories), fixed by the same line.
A body that says out for something else (not the Vec local, not a field) would meet the buffer parameter of that name: it is not the form, the old refusal stands.

#11112 (the last commit): a list that is looped over gets its element type, or the fn is refused

let classes = [64, 128, 256, 512]; for &c in &classes { lengths.push(c); } converted to const classes = [64, 128, 256, 512];, an untyped array that t27c reads as a tuple and a for cannot walk. The merge base writes the same constant for any fn with that pattern (fn total() -> u32 { let classes = [64, 128, 256]; .. for &c in &classes .. }); uniform_lengths was the first fn to reach it from a refusal, which is why padding_length_distribution_guard.rs flipped from a vacuous pass to BLOCKED.

let_array: when the element type is unknown and the list is written into or looped over (array_iterated: by name, by reference, .iter()), the fn is refused (R_VECTYPE, "a vec! or array literal whose element type or length t27o cannot tell"), and the token is now the list's name (classes, not its bracket). A list only indexed by number, or handed to a call, stays what it was.
put_vec_elem has a new rung, array_push_elem: in a fn in the Vec out-buffer form, a list whose loop binding is pushed as it is to the result Vec (loop_pushes: v.push(c), v the result local) has the Vec's element type, because what is pushed has it. v.push(c * 2) tells t27o nothing about c and is not guessed: it is refused.
Result on the file: const classes : [4]usize = [64, 128, 256, 512]; and for c in classes. The spec t27o writes for padding_length_distribution_guard.rs passes t27b test (1 passed) and t27c test-report (1 of 1, 10 runtime asserts, vacuous 0) instead of BLOCKED.
Tests first (four failed on the previous head): the pinned array-local test (its nums loop now refuses; the old comment said such a list "is left as it was"), the padding shape, the negative controls (a list looped over by reference, by .iter() and by value is refused naming the list; c * 2 is refused; a suffixed list is typed by its first element; a list only indexed by number is unchanged), and a direct test of fn_params_open, loop_pushes and array_push_elem. The refusal of a written list had named [; it names the list now.
The 17 files of the replay were converted again with the converter of the previous head and with this one: 16 are byte-identical, the 17th is the padding file (the typed array). The effect on the whole corpus of refusing an untyped list that is looped over (a fn that converted into an invalid spec before) is not measured, see "Not run".

Found, not fixed (filed as #11111 and #11113)

[bug] t27o: a caller of an out-buffer String fn is converted to a call without the buffer #11111: callers of an out-buffer String fn are converted to a call without the buffer: fn caller(n: u32) -> usize { tag(n).len() } becomes return tag(n).len; while tag takes (n, out). Reproduced on the merge base. This form refuses its callers instead.
t27o: the oracle of regex_lite_global never samples a match, so every recorded count is 0 #11113: the oracle of regex_lite_global is weak: its samples never make the needle match, so every recorded count is 0.

Checks

t27b test specs/tools/t27o/rust_convert.t27: 353 passed (333 before: 9 body and oracle tests and 8 direct tests for the Vec form, 2 body tests and 1 direct test for [bug] t27o: a read-only integer array local iterated by reference converts to an untyped array that t27c rejects #11112; every group failed first on the head before it). t27c test-report: vacuous passes 0 of 353. t27c coverage: 417 of 417 fns (149 of 149 for rust_std.t27, untouched, 51 of 51 pass).
Negative controls: a returned Vec that is read (len), inserted into, assigned into by index, sorted, popped, borrowed or rebound; a second Vec; a returned vec![n] or clone; a push whose own argument reads the Vec; a parameter already named out or a body that says out; elements that are String, tuple or char; collect, a vec! literal, vec![0u8; 4] and Option<Vec<u8>> results; a scratch Vec that is not returned; a call by name, by method and by path.
t27c gen-rust of the spec is byte-identical to the committed gen/rust/tools/t27o/rust_convert.rs, which compiles with rustc; zig_convert.t27 and spice_convert.t27 generate byte-identical Rust.
t27c seal --save for rust_convert (353 of 353 under Zig); --verify MATCH on rust_convert, rust_std, zig_convert, spice_convert, zig_std, ledger, golden, install, service (no dependent seal re-saved; spice_convert.t27 untouched).
python3 -I tools/dupe_scan.py: 738 of 15,920 bodies in 218 groups at the merge base, 738 of 15,941 at the head, the same 15 findings (none new); t27c dupes --name for each new helper: found only in this spec. The spec has 0 non-ASCII bytes and 0 home-directory literals.
Own-language gate, compiled from the merge base: 3 files (spec 970 / 28, generated Rust 561 / 21, seal 9 / 9 added / deleted lines), hand-written foreign lines added 0; the gate prints nothing and exits 0.

Not run

The replays of bootstrap/src/compiler.rs, service.rs, cli/tri/src/gates.rs (three of the five big files; their conversion changed) and of the four zero-parameter String files of the comma fix (trios-ui/rings/UR-04/src/lib.rs, archive/vibee-browser/src-tauri/src/main.rs in trinity and trinity-fpga, 999-multibots-telegraf/rust/rings/GOLD-RING-PR00/src/replicate.rs).
A sweep of the five repositories after the [bug] t27o: a read-only integer array local iterated by reference converts to an untyped array that t27c rejects #11112 commit: how many fns that converted into an untyped looped-over list are refused now is not measured (the 17 files above are the only check). Any sweep or replay on the rebased head itself, and cargo build -p tri with the real tri t27o --check (the driver I compiled with rustc makes the same calls as cmd_t27o; no hand Rust was added).
t27c suite, ./scripts/tri test, the CI workflows beyond the gates listed, tools/check_devhome_paths.py on a full tree (this checkout is sparse; the three changed files hold none), docs/reports/t27b_expectations.json and every ledger.
The commit hooks printed "no tri binary is built", so their gates did not run locally; each commit message carries Closes #11073 and the co-author line.

Draft; not set to auto-merge.

🤖 Generated with Claude Code

Whole-corpus sweep (added before merge)

All 1,388 tracked Rust files of the five repositories, merge-base converter against this head, one job at a time, 30 s cap per file, both ran every file to completion (rc 0, no timeouts), five big files included. Not the real tri binary: both converters compiled with rustc -O2 into the same per-file driver.

Measure	Merge base	This head	Change
fns seen	19,392	19,392	0
fns converted	1,820	1,824	+4
tests converted (of 9,230 seen)	619	619	0
first refusal `type Vec`	1,312	1,289	-23
first refusal, the new array refusal	0	11	+11
first refusal, a Vec result not built only by push	0	4	+4
first refusal, a call of an out-buffer fn	0	1	+1

30 of 1,388 output specs differ; 4 differ in code (all gains: bitstream.rs, jepa_t_ingest/src/lib.rs, padding_length_distribution_guard.rs, trios-phd/src/main.rs), 26 only in a refusal comment line. Fns that dropped from converted to refused: 0. Files from pass to not-pass: 0 (end-to-end replay of the four code-changing files: bitstream.rs 0 -> 1 tests, padding_length_distribution_guard.rs 0 -> 1 test, trios-phd 3 -> 4 tests, jepa_t_ingest the same existing failure oracle_token_to_q15). The 11 fns now refused by the array rule were already refused for a later reason. The payoff is small, as the measurement above says: 4 fns.

…form -- a last out: []T, the result the count, the oracle checks the written prefix (Closes #11073) Step 2 of #10546. A fn whose result is Vec<T> (T an integer, float or bool type), made as one local by Vec::new() / Vec::with_capacity(n) / vec![], changed only by push(e) and extend_from_slice(s) and handed back, takes the caller's buffer as a last out: []T and returns how many elements the Rust fn returns. An element past out.len is counted, not written, as the String form counts bytes; with_capacity(n) is a hint, read and dropped. A read of the Vec, any other change, a second Vec, any other result and a call of a fn in this form refuse, each with a reason that names the use (R_VECOUT, R_VECCALL). The oracle records the count and every element of the written prefix in a 256-element buffer and the same count in a zero-length one. Tests first (nine failed on the merge base), a negative control per rule, a direct test for every new fn (coverage 100 %). gen/rust/tools/t27o/rust_convert.rs is t27c gen-rust of the spec. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

…16 MiB, a Vec that never stops growing is skipped; a fn with no parameter takes the buffer without a comma; a body that says out for something else is not the form (Closes #11073) Found by the first replay of --check on a real file: crates/trios-phd regex_lite_global (s[i..].find(needle) with the sampled needle "") pushes for ever, and its harness ate the memory of the machine. A Vec result is an input-dependent allocation, so the harness now carries t27o_guard (VEC_GUARD) when some fn is in the Vec out-buffer form: the original runs on a thread with 3 seconds and a counting global allocator that parks the thread at 16 MiB; the sample is skipped like one on which Rust panics, two such samples end the sampling of that fn, and the cost of a skipped sample is a bounded 16 MiB. The same replay found padding_length_distribution_guard uniform_lengths() written as uniform_lengths(, t27o_o0[0 .. 256]) -- the comma was written before the buffer even with no argument (the String form had the same latent defect). And a body that says out for something else (not the Vec local, not a field) would meet the buffer parameter of that name: it is not the form, the old refusal stands. Tests first (each failed before its fix): a zero-parameter call, the guard in the harness and its absence for a String result, has_vecfn, three out-name rows. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

…ust takes the state by &mut and compiles (Closes #11073) The read-only form of the previous commit made t27c gen-rust type the state as impl AsRef<[usize]>, which moves the &mut [usize] of oracle() (E0382). The generated file is now compiled with rustc (the driver of cmd_t27o's calls) before every commit. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

…er -- 350 of 350 tests, vacuous 0 (Closes #11073) t27c seal --save of specs/tools/t27o/rust_convert.t27; --verify MATCHes on rust_convert, rust_std, zig_convert, spice_convert, zig_std, ledger, golden, install and service, so no dependent seal was re-saved. gen/rust/tools/t27o/rust_convert.rs is byte-identical to t27c gen-rust of the spec. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

github-actions · 2026-10-19T17:25:40Z

PR Dashboard

Generated at: 2026-10-19 17:25:29 UTC

t27o: a fn returning Vec<T> built by push converts in the out-buffer form -- a last out: []T, the result the count, the oracle checks the written prefix (Closes #11073, Closes #11112) #11109 t27o: a fn returning Vec built by push converts in the out-buffer form -- a last out: []T, the result the count, the oracle checks the written prefix (Closes #11073) | Created: 2026-10-19 | Updated: 2026-10-19 | 25 failing / 36 total
skills: record the night of 2026-10-19 six traps with tri learn (Closes #11106) #11108 skills: record the night of 2026-10-19 six traps with tri learn (Closes #11106) | Created: 2026-10-19 | Updated: 2026-10-19 | 22 failing / 29 total
widgets: canvas draw-list protocol and fifteen more native pages in .t27 #11107 widgets: canvas draw-list protocol and fifteen more native pages in .t27 | Created: 2026-10-19 | Updated: 2026-10-19 | 20 failing / 36 total
seals: ml_host_io_model gen_hash_verilog follows the gen-verilog assignment-target fix (Closes #11102) #11105 seals: ml_host_io_model gen_hash_verilog follows the gen-verilog assignment-target fix (Closes #11102) | Created: 2026-10-19 | Updated: 2026-10-19 | 12 failing / 31 total
spec(gen_w861): port scripts/gen_w861.py to specs/port/scripts/gen_w861.t27 #11093 spec(gen_w861): port scripts/gen_w861.py to specs/port/scripts/gen_w861.t27 | Created: 2026-10-19 | Updated: 2026-10-19 | 3 failing / 36 total
spec(gen_w862): port scripts/gen_w862.py to specs/port/scripts/gen_w862.t27 #11091 spec(gen_w862): port scripts/gen_w862.py to specs/port/scripts/gen_w862.t27 | Created: 2026-10-19 | Updated: 2026-10-19 | 3 failing / 36 total
spec(gen_w863): port scripts/gen_w863.py to specs/port/scripts/gen_w863.t27 #11090 spec(gen_w863): port scripts/gen_w863.py to specs/port/scripts/gen_w863.t27 | Created: 2026-10-19 | Updated: 2026-10-19 | 3 failing / 36 total
spec(gen_w864): port scripts/gen_w864.py to specs/port/scripts/gen_w864.t27 #11089 spec(gen_w864): port scripts/gen_w864.py to specs/port/scripts/gen_w864.t27 | Created: 2026-10-19 | Updated: 2026-10-19 | 3 failing / 36 total
spec(gen_w865): port scripts/gen_w865.py to specs/port/scripts/gen_w865.t27 #11088 spec(gen_w865): port scripts/gen_w865.py to specs/port/scripts/gen_w865.t27 | Created: 2026-10-19 | Updated: 2026-10-19 | 3 failing / 36 total
spec(gen_w866): port scripts/gen_w866.py to specs/port/scripts/gen_w866.t27 #11087 spec(gen_w866): port scripts/gen_w866.py to specs/port/scripts/gen_w866.t27 | Created: 2026-10-19 | Updated: 2026-10-19 | 3 failing / 36 total
spec(gen_w867): port scripts/gen_w867.py to specs/port/scripts/gen_w867.t27 #11086 spec(gen_w867): port scripts/gen_w867.py to specs/port/scripts/gen_w867.t27 | Created: 2026-10-19 | Updated: 2026-10-19 | 3 failing / 36 total
spec(gen_w868): port scripts/gen_w868.py to specs/port/scripts/gen_w868.t27 #11085 spec(gen_w868): port scripts/gen_w868.py to specs/port/scripts/gen_w868.t27 | Created: 2026-10-19 | Updated: 2026-10-19 | 5 failing / 35 total
fix(t27os): panes no longer inherit a Claude Code host session's environment #11029 fix(t27os): panes no longer inherit a Claude Code host session's environment | Created: 2026-10-19 | Updated: 2026-10-19 | 5 failing / 38 total
fix(t27os): CTL verbs match whole words; escaped keys and expect; agent check table #11015 fix(t27os): CTL verbs match whole words; escaped keys and expect; agent check table | Created: 2026-10-19 | Updated: 2026-10-19 | 5 failing / 38 total
feat(t27os): key manager, KEYS tab, per-pane profile, router decisions (Refs #10782) #10977 feat(t27os): key manager, KEYS tab, per-pane profile, router decisions (Refs #10782) | Created: 2026-10-19 | Updated: 2026-10-19 | 0 failing / 36 total
spec(ternary): trit codon -- perfect [4,2,3] Hamming code over GF(3), base-3 and wire forms (Closes #10969) #10970 spec(ternary): trit codon -- perfect [4,2,3] Hamming code over GF(3), base-3 and wire forms (Closes #10969) | Created: 2026-10-19 | Updated: 2026-10-19 | 5 failing / 38 total
fpga: T27 OS over UART on the AX7203 (feasibility + shell spec) (Refs #10782) #10949 fpga: T27 OS over UART on the AX7203 (feasibility + shell spec) (Refs #10782) | Created: 2026-10-19 | Updated: 2026-10-19 | 5 failing / 45 total
feat(igla): hslm_bwd_ptx kernel 3 -- split-K dW, 4.2-4.9x faster than kernel 2 at K=1782 (Closes #10895) #10914 feat(igla): hslm_bwd_ptx kernel 3 -- split-K dW, 4.2-4.9x faster than kernel 2 at K=1782 (Closes #10895) | Created: 2026-10-19 | Updated: 2026-10-19 | 8 failing / 46 total
feat(t27b): build record -- skip a build when nothing that feeds the object changed; --force #10890 feat(t27b): build record -- skip a build when nothing that feeds the object changed; --force | Created: 2026-10-19 | Updated: 2026-10-19 | 6 failing / 40 total
igla cell 25 published: the whole-model HSLM trained by t27c --backend ml is on Hugging Face (Closes #10833) #10836 igla cell 25 published: the whole-model HSLM trained by t27c --backend ml is on Hugging Face (Closes #10833) | Created: 2026-10-19 | Updated: 2026-10-19 | 6 failing / 40 total
spec(port): land the CR-CHAT-03 wave 3 batch -- 23 targets (Refs #8883) #10802 spec(port): land the CR-CHAT-03 wave 3 batch -- 23 targets (Refs #8883) | Created: 2026-10-19 | Updated: 2026-10-19 | 3 failing / 37 total
feat(t27f): T27 OS home -- the 27-petal triangle menu as one spec #10783 feat(t27f): T27 OS home -- the 27-petal triangle menu as one spec | Created: 2026-10-19 | Updated: 2026-10-19 | 3 failing / 37 total
queen: key_credit.t27 -- a leased contributor key earns credit only for .t27 that landed (Closes #10589) #10591 queen: key_credit.t27 -- a leased contributor key earns credit only for .t27 that landed (Closes #10589) | Created: 2026-10-18 | Updated: 2026-10-19 | 1 failing / 38 total
claw: criteria.t27 + claw/criteria check -- commands settle the criteria they can (Closes #10549) #10572 claw: criteria.t27 + claw/criteria check -- commands settle the criteria they can (Closes #10549) | Created: 2026-10-18 | Updated: 2026-10-19 | 1 failing / 31 total
queen: key_vault.t27 -- one key store, every consumer leases, review lanes first (Closes #10564) #10567 queen: key_vault.t27 -- one key store, every consumer leases, review lanes first (Closes #10564) | Created: 2026-10-18 | Updated: 2026-10-19 | 1 failing / 38 total
claw: reel_still.t27 -- the TRI CLAW /mcp tool that renders one particle still #10548 claw: reel_still.t27 -- the TRI CLAW /mcp tool that renders one particle still | Created: 2026-10-18 | Updated: 2026-10-19 | 0 failing / 28 total
fix(t27c): seal --verify reads the committed gen file; compile refuses a hand-written output (draft, needs owner-approved-foreign) #10481 fix(t27c): seal --verify reads the committed gen file; compile refuses a hand-written output (draft, needs owner-approved-foreign) | Created: 2026-10-18 | Updated: 2026-10-18 | 2 failing / 46 total
fix(seal): refuse a seal file another spec owns, re-check the hash after the write, land it by rename #10464 fix(seal): refuse a seal file another spec owns, re-check the hash after the write, land it by rename | Created: 2026-10-18 | Updated: 2026-10-18 | 7 failing / 48 total
igla ptx v2: cp.async ring, 128x128xBK64 tile, prmt-decoded W; within 1.33x of CUTLASS s8 (Refs #10404) #10456 igla ptx v2: cp.async ring, 128x128xBK64 tile, prmt-decoded W; within 1.33x of CUTLASS s8 (Refs #10404) | Created: 2026-10-18 | Updated: 2026-10-18 | 6 failing / 46 total
check: refuse a module that declares a top-level name twice (Closes #10421) #10442 check: refuse a module that declares a top-level name twice (Closes #10421) | Created: 2026-10-18 | Updated: 2026-10-18 | 5 failing / 44 total
demo: claw/sarif alert on a spec that does not parse (DO NOT MERGE, Refs #10356) #10381 demo: claw/sarif alert on a spec that does not parse (DO NOT MERGE, Refs #10356) | Created: 2026-10-18 | Updated: 2026-10-18 | 6 failing / 33 total
t27b: repair crypto and rust_convert; distinguish PR ledger lag #9637 t27b: repair crypto and rust_convert; distinguish PR ledger lag | Created: 2026-10-15 | Updated: 2026-10-15 | 3 failing / 33 total
spec(port): share tree helpers across eight gen_w ports (Closes #9633) #9636 spec(port): share tree helpers across eight gen_w ports (Closes #9633) | Created: 2026-10-15 | Updated: 2026-10-17 | 3 failing / 33 total
Port fpga/verilog/gft_xorpercep_jtag.v (Verilog, 1 module) to specs/port/fpga/verilog/gft_xorpercep_jtag.t27 #9549 Port fpga/verilog/gft_xorpercep_jtag.v (Verilog, 1 module) to specs/port/fpga/verilog/gft_xorpercep_jtag.t27 | Created: 2026-10-14 | Updated: 2026-10-14 | 4 failing / 29 total
Port gHashTag/trios:crates/trios-git/src/stage.rs (Rust, 2 functions) to specs/port/trios/crates/trios-git/src/stage.t27 #9548 Port gHashTag/trios:crates/trios-git/src/stage.rs (Rust, 2 functions) to specs/port/trios/crates/trios-git/src/stage.t27 | Created: 2026-10-14 | Updated: 2026-10-14 | 4 failing / 29 total
Port gHashTag/trios:crates/trios-rainbow-bridge/src/bridge.rs (Rust, 5 functions) to specs/port/trios/crates/trios-rainb #9485 Port gHashTag/trios:crates/trios-rainbow-bridge/src/bridge.rs (Rust, 5 functions) to specs/port/trios/crates/trios-rainb | Created: 2026-10-14 | Updated: 2026-10-14 | 4 failing / 29 total
Port scripts/gen_w734.py (Python, 3 functions) to specs/port/scripts/gen_w734.t27 #9484 Port scripts/gen_w734.py (Python, 3 functions) to specs/port/scripts/gen_w734.t27 | Created: 2026-10-14 | Updated: 2026-10-14 | 4 failing / 29 total
Port scripts/gen_w726.py (Python, 3 functions) to specs/port/scripts/gen_w726.t27 #9463 Port scripts/gen_w726.py (Python, 3 functions) to specs/port/scripts/gen_w726.t27 | Created: 2026-10-14 | Updated: 2026-10-14 | 3 failing / 29 total
Port gHashTag/trinity:src/tri/sacred_alu.zig (Zig, 4 functions) to specs/port/trinity/src/tri/sacred_alu.t27 #9427 Port gHashTag/trinity:src/tri/sacred_alu.zig (Zig, 4 functions) to specs/port/trinity/src/tri/sacred_alu.t27 | Created: 2026-10-13 | Updated: 2026-10-13 | 5 failing / 29 total
Port gHashTag/trios:crates/trios-phd/src/bin/rag_search.rs (Rust, 3 functions) to specs/port/trios/crates/trios-phd/src/ #9426 Port gHashTag/trios:crates/trios-phd/src/bin/rag_search.rs (Rust, 3 functions) to specs/port/trios/crates/trios-phd/src/ | Created: 2026-10-13 | Updated: 2026-10-13 | 4 failing / 29 total
Port gHashTag/trinity:src/vibeec/spec_loader.zig (Zig, 2 functions) to specs/port/trinity/src/vibeec/spec_loader.t27 #9407 Port gHashTag/trinity:src/vibeec/spec_loader.zig (Zig, 2 functions) to specs/port/trinity/src/vibeec/spec_loader.t27 | Created: 2026-10-13 | Updated: 2026-10-13 | 4 failing / 29 total
Port gHashTag/trinity:src/background_agent/config.zig (Zig, 5 functions) to specs/port/trinity/src/background_agent/conf #9345 Port gHashTag/trinity:src/background_agent/config.zig (Zig, 5 functions) to specs/port/trinity/src/background_agent/conf | Created: 2026-10-13 | Updated: 2026-10-13 | 5 failing / 29 total
Port gHashTag/trios:crates/trios-cli/src/cmd/train.rs (Rust, 6 functions) to specs/port/trios/crates/trios-cli/src/cmd/t #9321 Port gHashTag/trios:crates/trios-cli/src/cmd/train.rs (Rust, 6 functions) to specs/port/trios/crates/trios-cli/src/cmd/t | Created: 2026-10-13 | Updated: 2026-10-13 | 4 failing / 29 total
Port gHashTag/trinity:fpga/openxc7-synth/spi_flash_master.v (Verilog, 1 module) to specs/port/trinity/fpga/openxc7-synth #9263 Port gHashTag/trinity:fpga/openxc7-synth/spi_flash_master.v (Verilog, 1 module) to specs/port/trinity/fpga/openxc7-synth | Created: 2026-10-12 | Updated: 2026-10-12 | 7 failing / 29 total
spec fix: 4 specs in specs/tri/net the reference blocks -- rewrite them in t27 (1 of 1) #9214 spec fix: 4 specs in specs/tri/net the reference blocks -- rewrite them in t27 (1 of 1) | Created: 2026-10-12 | Updated: 2026-10-12 | 5 failing / 29 total
Port gHashTag/trios:crates/trios-agent-harness/rings/AH-00/src/lib.rs (Rust, 3 functions) to specs/port/trios/crates/tri #9200 Port gHashTag/trios:crates/trios-agent-harness/rings/AH-00/src/lib.rs (Rust, 3 functions) to specs/port/trios/crates/tri | Created: 2026-10-12 | Updated: 2026-10-12 | 6 failing / 29 total
Port bootstrap/src/compiler.rs lines 30977-31050 (Rust, 6 functions) to specs/port/bootstrap/src/compiler/LinkSegment_te #9178 Port bootstrap/src/compiler.rs lines 30977-31050 (Rust, 6 functions) to specs/port/bootstrap/src/compiler/LinkSegment_te | Created: 2026-10-11 | Updated: 2026-10-12 | 6 failing / 29 total
Port gHashTag/trinity:src/storm/links/vibee.zig (Zig, 1 function) to specs/port/trinity/src/storm/links/vibee.t27 #9177 Port gHashTag/trinity:src/storm/links/vibee.zig (Zig, 1 function) to specs/port/trinity/src/storm/links/vibee.t27 | Created: 2026-10-11 | Updated: 2026-10-12 | 5 failing / 32 total
Port scripts/gen_w706.py (Python, 3 functions) to specs/port/scripts/gen_w706.t27 #9150 Port scripts/gen_w706.py (Python, 3 functions) to specs/port/scripts/gen_w706.t27 | Created: 2026-10-11 | Updated: 2026-10-12 | 7 failing / 30 total
spec fix: 1 specs in specs/queen the reference blocks -- rewrite them in t27 (1 of 1) #8879 spec fix: 1 specs in specs/queen the reference blocks -- rewrite them in t27 (1 of 1) | Created: 2026-10-11 | Updated: 2026-10-18 | 4 failing / 30 total

Summary

Status	Count
Total Open PRs	50
PRs with Failing Checks	44
PRs with All Checks Green	6
READY	2
FAILING	44
PENDING	0
NO CHECKS YET	0

These columns do not partition: 2 + 44 + 0 + 0 = 46, and there are 50 open PRs. A PR is being counted twice or not at all.

Seal Status

⚠️ STALE -- sha256(compiler.rs)=e34e1b174c96 != manifest seal=87e5cbd3ad94.
The committed NMSE numbers were certified against an older compiler.rs.
Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

…ming the list, and one whose binding is pushed to the Vec result takes the Vec's element type (Closes #11112) let classes = [64, 128, 256, 512]; for &c in &classes { lengths.push(c); } converted to an untyped array constant, which t27c reads as a tuple and a for cannot walk: padding_length_distribution_guard.rs went from a vacuous pass to BLOCKED once uniform_lengths started to convert (#11073). The merge base writes the same untyped constant for any fn with that pattern. - let_array: when the element type is unknown and the list is written into or looped over (array_iterated: by name, by reference, .iter()), the fn is refused (R_VECTYPE), the token now the list's name, not its bracket. A list only indexed by number or handed to a call stays what it was. - put_vec_elem has a new rung, array_push_elem: in a fn in the Vec out-buffer form, a list whose loop binding is pushed as it is to the result Vec (loop_pushes) has the Vec's element type -- what is pushed has it. v.push(c * 2) tells nothing and is not guessed. Tests first (four failed on the previous head): the pinned array-local test, the padding shape, the negative controls (by reference, .iter(), by value, c * 2 refused; a suffixed list typed; a list only indexed by number unchanged), and a direct test of the three new fns. The spec t27o writes for the padding file now passes t27c test-report (1 of 1, 10 runtime asserts, vacuous 0) instead of BLOCKED. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

github-actions · 2026-10-19T17:38:55Z

PR Dashboard

Generated at: 2026-10-19 17:38:46 UTC

t27o: a fn returning Vec<T> built by push converts in the out-buffer form -- a last out: []T, the result the count, the oracle checks the written prefix (Closes #11073, Closes #11112) #11109 t27o: a fn returning Vec built by push converts in the out-buffer form -- a last out: []T, the result the count, the oracle checks the written prefix (Closes #11073, Closes #11112) | Created: 2026-10-19 | Updated: 2026-10-19 | 12 failing / 35 total
skills: record the night of 2026-10-19 six traps with tri learn (Closes #11106) #11108 skills: record the night of 2026-10-19 six traps with tri learn (Closes #11106) | Created: 2026-10-19 | Updated: 2026-10-19 | 5 failing / 29 total
spec(gen_w861): port scripts/gen_w861.py to specs/port/scripts/gen_w861.t27 #11093 spec(gen_w861): port scripts/gen_w861.py to specs/port/scripts/gen_w861.t27 | Created: 2026-10-19 | Updated: 2026-10-19 | 3 failing / 36 total
spec(gen_w862): port scripts/gen_w862.py to specs/port/scripts/gen_w862.t27 #11091 spec(gen_w862): port scripts/gen_w862.py to specs/port/scripts/gen_w862.t27 | Created: 2026-10-19 | Updated: 2026-10-19 | 3 failing / 36 total
spec(gen_w863): port scripts/gen_w863.py to specs/port/scripts/gen_w863.t27 #11090 spec(gen_w863): port scripts/gen_w863.py to specs/port/scripts/gen_w863.t27 | Created: 2026-10-19 | Updated: 2026-10-19 | 3 failing / 36 total
spec(gen_w864): port scripts/gen_w864.py to specs/port/scripts/gen_w864.t27 #11089 spec(gen_w864): port scripts/gen_w864.py to specs/port/scripts/gen_w864.t27 | Created: 2026-10-19 | Updated: 2026-10-19 | 3 failing / 36 total
spec(gen_w865): port scripts/gen_w865.py to specs/port/scripts/gen_w865.t27 #11088 spec(gen_w865): port scripts/gen_w865.py to specs/port/scripts/gen_w865.t27 | Created: 2026-10-19 | Updated: 2026-10-19 | 3 failing / 36 total
spec(gen_w866): port scripts/gen_w866.py to specs/port/scripts/gen_w866.t27 #11087 spec(gen_w866): port scripts/gen_w866.py to specs/port/scripts/gen_w866.t27 | Created: 2026-10-19 | Updated: 2026-10-19 | 3 failing / 36 total
spec(gen_w867): port scripts/gen_w867.py to specs/port/scripts/gen_w867.t27 #11086 spec(gen_w867): port scripts/gen_w867.py to specs/port/scripts/gen_w867.t27 | Created: 2026-10-19 | Updated: 2026-10-19 | 3 failing / 36 total
spec(gen_w868): port scripts/gen_w868.py to specs/port/scripts/gen_w868.t27 #11085 spec(gen_w868): port scripts/gen_w868.py to specs/port/scripts/gen_w868.t27 | Created: 2026-10-19 | Updated: 2026-10-19 | 16 failing / 35 total
fix(t27os): panes no longer inherit a Claude Code host session's environment #11029 fix(t27os): panes no longer inherit a Claude Code host session's environment | Created: 2026-10-19 | Updated: 2026-10-19 | 5 failing / 38 total
fix(t27os): CTL verbs match whole words; escaped keys and expect; agent check table #11015 fix(t27os): CTL verbs match whole words; escaped keys and expect; agent check table | Created: 2026-10-19 | Updated: 2026-10-19 | 5 failing / 38 total
feat(t27os): key manager, KEYS tab, per-pane profile, router decisions (Refs #10782) #10977 feat(t27os): key manager, KEYS tab, per-pane profile, router decisions (Refs #10782) | Created: 2026-10-19 | Updated: 2026-10-19 | 0 failing / 36 total
spec(ternary): trit codon -- perfect [4,2,3] Hamming code over GF(3), base-3 and wire forms (Closes #10969) #10970 spec(ternary): trit codon -- perfect [4,2,3] Hamming code over GF(3), base-3 and wire forms (Closes #10969) | Created: 2026-10-19 | Updated: 2026-10-19 | 5 failing / 38 total
fpga: T27 OS over UART on the AX7203 (feasibility + shell spec) (Refs #10782) #10949 fpga: T27 OS over UART on the AX7203 (feasibility + shell spec) (Refs #10782) | Created: 2026-10-19 | Updated: 2026-10-19 | 5 failing / 45 total
feat(igla): hslm_bwd_ptx kernel 3 -- split-K dW, 4.2-4.9x faster than kernel 2 at K=1782 (Closes #10895) #10914 feat(igla): hslm_bwd_ptx kernel 3 -- split-K dW, 4.2-4.9x faster than kernel 2 at K=1782 (Closes #10895) | Created: 2026-10-19 | Updated: 2026-10-19 | 8 failing / 46 total
feat(t27b): build record -- skip a build when nothing that feeds the object changed; --force #10890 feat(t27b): build record -- skip a build when nothing that feeds the object changed; --force | Created: 2026-10-19 | Updated: 2026-10-19 | 6 failing / 40 total
igla cell 25 published: the whole-model HSLM trained by t27c --backend ml is on Hugging Face (Closes #10833) #10836 igla cell 25 published: the whole-model HSLM trained by t27c --backend ml is on Hugging Face (Closes #10833) | Created: 2026-10-19 | Updated: 2026-10-19 | 6 failing / 40 total
spec(port): land the CR-CHAT-03 wave 3 batch -- 23 targets (Refs #8883) #10802 spec(port): land the CR-CHAT-03 wave 3 batch -- 23 targets (Refs #8883) | Created: 2026-10-19 | Updated: 2026-10-19 | 3 failing / 37 total
feat(t27f): T27 OS home -- the 27-petal triangle menu as one spec #10783 feat(t27f): T27 OS home -- the 27-petal triangle menu as one spec | Created: 2026-10-19 | Updated: 2026-10-19 | 3 failing / 37 total
queen: key_credit.t27 -- a leased contributor key earns credit only for .t27 that landed (Closes #10589) #10591 queen: key_credit.t27 -- a leased contributor key earns credit only for .t27 that landed (Closes #10589) | Created: 2026-10-18 | Updated: 2026-10-19 | 1 failing / 38 total
claw: criteria.t27 + claw/criteria check -- commands settle the criteria they can (Closes #10549) #10572 claw: criteria.t27 + claw/criteria check -- commands settle the criteria they can (Closes #10549) | Created: 2026-10-18 | Updated: 2026-10-19 | 1 failing / 31 total
queen: key_vault.t27 -- one key store, every consumer leases, review lanes first (Closes #10564) #10567 queen: key_vault.t27 -- one key store, every consumer leases, review lanes first (Closes #10564) | Created: 2026-10-18 | Updated: 2026-10-19 | 1 failing / 38 total
claw: reel_still.t27 -- the TRI CLAW /mcp tool that renders one particle still #10548 claw: reel_still.t27 -- the TRI CLAW /mcp tool that renders one particle still | Created: 2026-10-18 | Updated: 2026-10-19 | 0 failing / 28 total
fix(t27c): seal --verify reads the committed gen file; compile refuses a hand-written output (draft, needs owner-approved-foreign) #10481 fix(t27c): seal --verify reads the committed gen file; compile refuses a hand-written output (draft, needs owner-approved-foreign) | Created: 2026-10-18 | Updated: 2026-10-18 | 2 failing / 46 total
fix(seal): refuse a seal file another spec owns, re-check the hash after the write, land it by rename #10464 fix(seal): refuse a seal file another spec owns, re-check the hash after the write, land it by rename | Created: 2026-10-18 | Updated: 2026-10-18 | 7 failing / 48 total
igla ptx v2: cp.async ring, 128x128xBK64 tile, prmt-decoded W; within 1.33x of CUTLASS s8 (Refs #10404) #10456 igla ptx v2: cp.async ring, 128x128xBK64 tile, prmt-decoded W; within 1.33x of CUTLASS s8 (Refs #10404) | Created: 2026-10-18 | Updated: 2026-10-18 | 6 failing / 46 total
check: refuse a module that declares a top-level name twice (Closes #10421) #10442 check: refuse a module that declares a top-level name twice (Closes #10421) | Created: 2026-10-18 | Updated: 2026-10-18 | 5 failing / 44 total
demo: claw/sarif alert on a spec that does not parse (DO NOT MERGE, Refs #10356) #10381 demo: claw/sarif alert on a spec that does not parse (DO NOT MERGE, Refs #10356) | Created: 2026-10-18 | Updated: 2026-10-18 | 6 failing / 33 total
t27b: repair crypto and rust_convert; distinguish PR ledger lag #9637 t27b: repair crypto and rust_convert; distinguish PR ledger lag | Created: 2026-10-15 | Updated: 2026-10-15 | 3 failing / 33 total
spec(port): share tree helpers across eight gen_w ports (Closes #9633) #9636 spec(port): share tree helpers across eight gen_w ports (Closes #9633) | Created: 2026-10-15 | Updated: 2026-10-17 | 3 failing / 33 total
Port fpga/verilog/gft_xorpercep_jtag.v (Verilog, 1 module) to specs/port/fpga/verilog/gft_xorpercep_jtag.t27 #9549 Port fpga/verilog/gft_xorpercep_jtag.v (Verilog, 1 module) to specs/port/fpga/verilog/gft_xorpercep_jtag.t27 | Created: 2026-10-14 | Updated: 2026-10-14 | 4 failing / 29 total
Port gHashTag/trios:crates/trios-git/src/stage.rs (Rust, 2 functions) to specs/port/trios/crates/trios-git/src/stage.t27 #9548 Port gHashTag/trios:crates/trios-git/src/stage.rs (Rust, 2 functions) to specs/port/trios/crates/trios-git/src/stage.t27 | Created: 2026-10-14 | Updated: 2026-10-14 | 4 failing / 29 total
Port gHashTag/trios:crates/trios-rainbow-bridge/src/bridge.rs (Rust, 5 functions) to specs/port/trios/crates/trios-rainb #9485 Port gHashTag/trios:crates/trios-rainbow-bridge/src/bridge.rs (Rust, 5 functions) to specs/port/trios/crates/trios-rainb | Created: 2026-10-14 | Updated: 2026-10-14 | 4 failing / 29 total
Port scripts/gen_w734.py (Python, 3 functions) to specs/port/scripts/gen_w734.t27 #9484 Port scripts/gen_w734.py (Python, 3 functions) to specs/port/scripts/gen_w734.t27 | Created: 2026-10-14 | Updated: 2026-10-14 | 4 failing / 29 total
Port scripts/gen_w726.py (Python, 3 functions) to specs/port/scripts/gen_w726.t27 #9463 Port scripts/gen_w726.py (Python, 3 functions) to specs/port/scripts/gen_w726.t27 | Created: 2026-10-14 | Updated: 2026-10-14 | 3 failing / 29 total
Port gHashTag/trinity:src/tri/sacred_alu.zig (Zig, 4 functions) to specs/port/trinity/src/tri/sacred_alu.t27 #9427 Port gHashTag/trinity:src/tri/sacred_alu.zig (Zig, 4 functions) to specs/port/trinity/src/tri/sacred_alu.t27 | Created: 2026-10-13 | Updated: 2026-10-13 | 5 failing / 29 total
Port gHashTag/trios:crates/trios-phd/src/bin/rag_search.rs (Rust, 3 functions) to specs/port/trios/crates/trios-phd/src/ #9426 Port gHashTag/trios:crates/trios-phd/src/bin/rag_search.rs (Rust, 3 functions) to specs/port/trios/crates/trios-phd/src/ | Created: 2026-10-13 | Updated: 2026-10-13 | 4 failing / 29 total
Port gHashTag/trinity:src/vibeec/spec_loader.zig (Zig, 2 functions) to specs/port/trinity/src/vibeec/spec_loader.t27 #9407 Port gHashTag/trinity:src/vibeec/spec_loader.zig (Zig, 2 functions) to specs/port/trinity/src/vibeec/spec_loader.t27 | Created: 2026-10-13 | Updated: 2026-10-13 | 4 failing / 29 total
Port gHashTag/trinity:src/background_agent/config.zig (Zig, 5 functions) to specs/port/trinity/src/background_agent/conf #9345 Port gHashTag/trinity:src/background_agent/config.zig (Zig, 5 functions) to specs/port/trinity/src/background_agent/conf | Created: 2026-10-13 | Updated: 2026-10-13 | 5 failing / 29 total
Port gHashTag/trios:crates/trios-cli/src/cmd/train.rs (Rust, 6 functions) to specs/port/trios/crates/trios-cli/src/cmd/t #9321 Port gHashTag/trios:crates/trios-cli/src/cmd/train.rs (Rust, 6 functions) to specs/port/trios/crates/trios-cli/src/cmd/t | Created: 2026-10-13 | Updated: 2026-10-13 | 4 failing / 29 total
Port gHashTag/trinity:fpga/openxc7-synth/spi_flash_master.v (Verilog, 1 module) to specs/port/trinity/fpga/openxc7-synth #9263 Port gHashTag/trinity:fpga/openxc7-synth/spi_flash_master.v (Verilog, 1 module) to specs/port/trinity/fpga/openxc7-synth | Created: 2026-10-12 | Updated: 2026-10-12 | 7 failing / 29 total
spec fix: 4 specs in specs/tri/net the reference blocks -- rewrite them in t27 (1 of 1) #9214 spec fix: 4 specs in specs/tri/net the reference blocks -- rewrite them in t27 (1 of 1) | Created: 2026-10-12 | Updated: 2026-10-12 | 5 failing / 29 total
Port gHashTag/trios:crates/trios-agent-harness/rings/AH-00/src/lib.rs (Rust, 3 functions) to specs/port/trios/crates/tri #9200 Port gHashTag/trios:crates/trios-agent-harness/rings/AH-00/src/lib.rs (Rust, 3 functions) to specs/port/trios/crates/tri | Created: 2026-10-12 | Updated: 2026-10-12 | 6 failing / 29 total
Port bootstrap/src/compiler.rs lines 30977-31050 (Rust, 6 functions) to specs/port/bootstrap/src/compiler/LinkSegment_te #9178 Port bootstrap/src/compiler.rs lines 30977-31050 (Rust, 6 functions) to specs/port/bootstrap/src/compiler/LinkSegment_te | Created: 2026-10-11 | Updated: 2026-10-12 | 6 failing / 29 total
Port gHashTag/trinity:src/storm/links/vibee.zig (Zig, 1 function) to specs/port/trinity/src/storm/links/vibee.t27 #9177 Port gHashTag/trinity:src/storm/links/vibee.zig (Zig, 1 function) to specs/port/trinity/src/storm/links/vibee.t27 | Created: 2026-10-11 | Updated: 2026-10-12 | 5 failing / 32 total
Port scripts/gen_w706.py (Python, 3 functions) to specs/port/scripts/gen_w706.t27 #9150 Port scripts/gen_w706.py (Python, 3 functions) to specs/port/scripts/gen_w706.t27 | Created: 2026-10-11 | Updated: 2026-10-12 | 7 failing / 30 total
spec fix: 1 specs in specs/queen the reference blocks -- rewrite them in t27 (1 of 1) #8879 spec fix: 1 specs in specs/queen the reference blocks -- rewrite them in t27 (1 of 1) | Created: 2026-10-11 | Updated: 2026-10-18 | 4 failing / 30 total
Port gHashTag/trinity:src/vibeec/updated_codex.zig (Zig, 5 functions) to specs/port/trinity/src/vibeec/updated_codex.t27 #8874 Port gHashTag/trinity:src/vibeec/updated_codex.zig (Zig, 5 functions) to specs/port/trinity/src/vibeec/updated_codex.t27 | Created: 2026-10-11 | Updated: 2026-10-12 | 6 failing / 30 total
Port gHashTag/trios:crates/trios-chat/rings/CR-CHAT-02/src/ratchet_replay_window_guard.rs (Rust, 2 functions) to specs/p #8873 Port gHashTag/trios:crates/trios-chat/rings/CR-CHAT-02/src/ratchet_replay_window_guard.rs (Rust, 2 functions) to specs/p | Created: 2026-10-11 | Updated: 2026-10-12 | 9 failing / 31 total

Summary

Status	Count
Total Open PRs	50
PRs with Failing Checks	46
PRs with All Checks Green	4
READY	2
FAILING	46
PENDING	0
NO CHECKS YET	0

These columns do not partition: 2 + 46 + 0 + 0 = 48, and there are 50 open PRs. A PR is being counted twice or not at all.

Seal Status

⚠️ STALE -- sha256(compiler.rs)=e34e1b174c96 != manifest seal=87e5cbd3ad94.
The committed NMSE numbers were certified against an older compiler.rs.
Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

@max

…stack, in both orders, and each body literal beside a sample that holds it -- regex_lite_global records a match (Closes #11113) The oracle calls a fn with parameter i taking sample j + i. regex_lite_global(s, needle) of trios-phd was called with ("", "a"), ("a", "abc"), ("abc", "0") ... : the needle was never inside the haystack, so every count the harness recorded was 0, the oracle test asserted only the empty case, and a conversion that returned nothing for every input would have passed it. Its body has no string literal, so the literal samples of #9102 could not help. vectors now counts the direct texts of the fn (text_count, kept in the new PAIRS slot and cleared at the end). With two or more, derived_at gives each text parameter three fixed samples first, a text, a text inside it and the first text again (put_pair_sample: "abcabc", "abc", "abcabc"), so one call has the needle inside the haystack, the next the haystack second, and "abc" occurs twice in "abcabc" (offsets 0 and 3, the second ending at the end). Each string literal of the body then comes alone, wrapped in one space, alone, so the wrapped sample sits next to the literal in both orders. The sampler stays bounded: three fixed 6-byte texts, the DCAP of 64 samples a parameter (a test with thirty literals), and the guard of 3 s and 16 MiB of #11109, which still skips the empty needle. A fn with one text, or a text and a number, is sampled as before (negative control); a number never reads the count of texts. Evidence: the harness of crates/trios-phd/src/main.rs now records count 2 and the offsets 0 and 3 for regex_lite_global (rustc, run), the zero-length buffer answers 2, and t27b test of the converted spec passes 4 of 4 as before. Over the 1,388 files of the five repositories the converted specs are byte for byte the same; 5 oracle harnesses change (compiler.rs, gates.rs, ring-090 lib.rs, trios-phd main.rs, defense_gate lib.rs); the three that t27b judges (ring-090 25 of 25, trios-phd 4 of 4, defense_gate 4 of 4) pass before and after, the two it cannot judge (unsupported construct, @max) stay so. Tests first (the two oracle tests failed on the merge base, the negative control passed), a direct test for each new fn (is_text_kind, text_count, put_pair_sample), derived_at with the count set, the cap of 64 with pairs. Three tests that give the name table exactly one slot got one more: the state vector has one more slot (PAIRS). t27b test 362 of 362, t27c coverage 100 % (420 of 420), seal --save 362 of 362 vacuous 0, --verify MATCH on rust_convert, rust_std, zig_convert, spice_convert, zig_std, ledger, golden, install, service. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

@max

… match in the oracle of a two-text fn, refused receivers named (#11130) * t27o: a caller of an out-buffer String fn is refused with a reason that names the form -- R_STRCALL, as the Vec form does with R_VECCALL (Closes #11111) A fn whose result is an owned String lowers to the out-buffer form: a last out: []u8, the result the byte count. The Rust source calls it without that argument, so a caller came out as tag(n).len or const t = tag(n); -- a call with the buffer missing, which t27c rejects at best and at worst reads as another meaning of the same arity. The converter must emit valid output or refuse (#10976). deps now treats a fn in the String form (K_STRFN, registered before its body is read) the way it treats one in the Vec form: a free fn is hit by a call, a method by a call after a dot or a path through its owner or Self, and the caller is refused with R_STRCALL, whose reason names the form and whose token is the callee. The buffer is not guessed: a caller that owned one would have to state a length the source never states. A fn outside the form keeps its old reason (uses refused), the split chain, which names the fn it calls on each piece without calling it, still converts. Tests first (two of the three new convert tests failed on the merge base, the third is the negative control and passed): a free fn in either declaration order, a method called as x.m() and S::m(x), a caller of a String fn outside the form, and a direct deps test with K_STRFN names (call, bare name, declaration, method, path, wrong owner, Self). The reason table test has the new row. Evidence: t27b test 357 of 357; t27c coverage 100 % (417 of 417); t27c seal --save 357 of 357, vacuous 0, --verify MATCH on rust_convert, rust_std, zig_convert, spice_convert, zig_std, ledger, golden, install, service. The converter over all 1,388 tracked Rust files of the five repositories changes no output: 13 fns in 9 files are in the String form and none is called in its own file. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com> * t27o: the oracle of a fn with two texts samples a needle inside a haystack, in both orders, and each body literal beside a sample that holds it -- regex_lite_global records a match (Closes #11113) The oracle calls a fn with parameter i taking sample j + i. regex_lite_global(s, needle) of trios-phd was called with ("", "a"), ("a", "abc"), ("abc", "0") ... : the needle was never inside the haystack, so every count the harness recorded was 0, the oracle test asserted only the empty case, and a conversion that returned nothing for every input would have passed it. Its body has no string literal, so the literal samples of #9102 could not help. vectors now counts the direct texts of the fn (text_count, kept in the new PAIRS slot and cleared at the end). With two or more, derived_at gives each text parameter three fixed samples first, a text, a text inside it and the first text again (put_pair_sample: "abcabc", "abc", "abcabc"), so one call has the needle inside the haystack, the next the haystack second, and "abc" occurs twice in "abcabc" (offsets 0 and 3, the second ending at the end). Each string literal of the body then comes alone, wrapped in one space, alone, so the wrapped sample sits next to the literal in both orders. The sampler stays bounded: three fixed 6-byte texts, the DCAP of 64 samples a parameter (a test with thirty literals), and the guard of 3 s and 16 MiB of #11109, which still skips the empty needle. A fn with one text, or a text and a number, is sampled as before (negative control); a number never reads the count of texts. Evidence: the harness of crates/trios-phd/src/main.rs now records count 2 and the offsets 0 and 3 for regex_lite_global (rustc, run), the zero-length buffer answers 2, and t27b test of the converted spec passes 4 of 4 as before. Over the 1,388 files of the five repositories the converted specs are byte for byte the same; 5 oracle harnesses change (compiler.rs, gates.rs, ring-090 lib.rs, trios-phd main.rs, defense_gate lib.rs); the three that t27b judges (ring-090 25 of 25, trios-phd 4 of 4, defense_gate 4 of 4) pass before and after, the two it cannot judge (unsupported construct, @max) stay so. Tests first (the two oracle tests failed on the merge base, the negative control passed), a direct test for each new fn (is_text_kind, text_count, put_pair_sample), derived_at with the count set, the cap of 64 with pairs. Three tests that give the name table exactly one slot got one more: the state vector has one more slot (PAIRS). t27b test 362 of 362, t27c coverage 100 % (420 of 420), seal --save 362 of 362 vacuous 0, --verify MATCH on rust_convert, rust_std, zig_convert, spice_convert, zig_std, ledger, golden, install, service. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com> * t27o: a refusal caused by a refused receiver names the receiver type, not `self` or `Self` -- 649 fns of the sweep can be followed to the type's own reason (Closes #11128) A method of a type the converter refuses (a struct with a heap field, an enum with data, a refused item) is refused with it, and its refusal line said "uses refused `self`" or "uses refused `Self`". Neither is the name of any refusal in the file, so a lookup by name found nothing and the sweep, which follows a named callee to its root (#11122), had to leave the cascade as it was: 696 fn of the 4,468 cascades. deps now refuses at the receiver (the `self` of the signature, or the `Self` of a body) with the impl type as the token, the line staying the receiver's: "uses refused `Foo` at line 3", and `Foo` has its own line, "refused struct Foo (line 1): type `String`". Nothing else changes: a static fn of the same impl that names neither the type nor `self` converts, the methods of a type that converts convert, a method that uses a refused method still names that method, and an impl of a type declared nowhere in the file stays "undeclared" (that check comes first). Tests first (the new convert test and the two new asserts of the direct deps test failed on the merge base): `&self`, a `Self` constructor, a method calling another, a refused enum, the free fn that takes the type, and the controls; one pinned line of an older test, `uses refused `Self``, now names the type. Measured on the 1,388 tracked Rust files of the five repositories (19,392 fn): the sweep's root attribution of #11127, replayed on the converted text (its numbers reproduced on the merge base: 4,531 cascade lines, 3,824 attributed, 696 fn and 707 items left, 570 + 86 of them self and Self), leaves 58 fn and 69 items unattributed after this change, none of them a receiver. Fns converted 1,824 before and after; none converted becomes refused, none refused becomes converted; 649 first refusals change their token and nothing else. What is left are names the file's own refusal lines do not carry: the type aliases the generic item refusal prints as "refused type" (47 fn), extern fns and an assert. t27b test 363 of 363, t27c coverage 100 % (420 of 420), seal --save 363 of 363 vacuous 0, --verify MATCH on rust_convert, rust_std, zig_convert, spice_convert, zig_std, ledger, golden, install, service. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com> --------- Co-authored-by: Claude <claude@anthropic.com> Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>

claude and others added 4 commits October 20, 2026 00:17

This was referenced Oct 19, 2026

[bug] t27o: a caller of an out-buffer String fn is converted to a call without the buffer #11111

Closed

t27o: the oracle of regex_lite_global never samples a match, so every recorded count is 0 #11113

Closed

github-actions Bot mentioned this pull request Oct 19, 2026

widgets: canvas draw-list protocol and fifteen more native pages in .t27 #11107

Merged

This was referenced Oct 19, 2026

skills: record the night of 2026-10-19 six traps with tri learn (Closes #11106) #11108

Merged

tri learn: a code fence closes only on a bare line as long as its opener (Closes #11110) #11114

Merged

gHashTag marked this pull request as ready for review October 19, 2026 17:50

gHashTag merged commit 2e2ca5f into master Oct 19, 2026
37 of 40 checks passed

gHashTag mentioned this pull request Oct 19, 2026

[EPIC] t27o overnight loop: dogfood on the roadmap repos, t27b-fast checks, refusal classes, competitors (feeds #10486 M2) #10522

Open

9 of 12 tasks

gHashTag mentioned this pull request Oct 19, 2026

t27o: converter trio -- callers of an out-buffer String fn refused, a match in the oracle of a two-text fn, refused receivers named #11130

Merged

gHashTag deleted the t27o/vec-out-buffer-a4 branch October 20, 2026 04:46

gHashTag mentioned this pull request Oct 20, 2026

docs: keep the t27o night report in the repository #11281

Merged

Navigation Menu

Repository navigation

t27o: a fn returning Vec<T> built by push converts in the out-buffer form -- a last out: []T, the result the count, the oracle checks the written prefix (Closes #11073, Closes #11112) - #11109

t27o: a fn returning Vec<T> built by push converts in the out-buffer form -- a last out: []T, the result the count, the oracle checks the written prefix (Closes #11073, Closes #11112)#11109
gHashTag merged 5 commits into
masterfrom
t27o/vec-out-buffer-a4

gHashTag commented Oct 19, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Oct 19, 2026

Uh oh!

github-actions Bot commented Oct 19, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

gHashTag commented Oct 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

The rule, and where it lives

Measure (the five repositories at the commits of step 1: t27 7496ed2, trios 483dd9ee, trinity 0770abb9, trinity-fpga 20bf4c418, 999-multibots-telegraf 6e5512881; 1,388 tracked .rs files, gen/ and target/ left out; the converter built from the generated Rust, one file per process, 60 s cap)

Replay of --check (rustc harness, run, t27b test of spec plus samples), measured on 46b3eff, the head before the rebase (the t27o files did not change on master since; the rebased gen file is byte-identical to t27c gen-rust of the spec)

Found by the replay, fixed here (each test written first and failing)

#11112 (the last commit): a list that is looped over gets its element type, or the fn is refused

Found, not fixed (filed as #11111 and #11113)

Checks

Not run

Whole-corpus sweep (added before merge)

Uh oh!

github-actions Bot commented Oct 19, 2026

PR Dashboard

Summary

Seal Status

Uh oh!

github-actions Bot commented Oct 19, 2026

PR Dashboard

Summary

Seal Status

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

gHashTag commented Oct 19, 2026 •

edited

Loading

Measure (the five repositories at the commits of step 1: t27 `7496ed2`, trios 483dd9ee, trinity 0770abb9, trinity-fpga 20bf4c418, 999-multibots-telegraf 6e5512881; 1,388 tracked `.rs` files, `gen/` and `target/` left out; the converter built from the generated Rust, one file per process, 60 s cap)

Replay of `--check` (rustc harness, run, `t27b test` of spec plus samples), measured on `46b3eff`, the head before the rebase (the t27o files did not change on master since; the rebased `gen` file is byte-identical to `t27c gen-rust` of the spec)