feat(ring-1): lex all 28 specs without errors [SEED-1] by gHashTag · Pull Request #9 · gHashTag/t27

gHashTag · 2026-04-04T09:49:53Z

Closes #4

Summary

All 27 spec files in specs/ now parse successfully with t27c
All 15 compiler spec files in compiler/ also parse successfully
Single file changed: bootstrap/src/compiler.rs (+273, -51)

Changes

Brace-style modules: module Name { ... } (was only module name;)
Arrow return types: fn foo() -> Type { ... }
Generic returns: Option<T>, [][]const u8, !void
var declarations: at module level
Keyword-style blocks: test name given ... when ... then ...
Block comments: /* ... */
Dotted fn names: Parser.new(...)
Error union returns: !void, !Type
Const qualifier on fn: fn foo() const { ... }

Test plan

27/27 spec files parse without errors
15/15 compiler spec files parse without errors
cargo test passes (0 regressions)
Ring-0 functionality preserved

φ² + 1/φ² = 3 | TRINITY

All spec files in specs/ and compiler/ now parse successfully with t27c. Fixes: - Brace-style module declarations (module Name { ... }) - Arrow return types (fn foo() -> Type) - Generic return types (Option<T>) - Slice/const return types ([]const u8, [][]const u8) - Error union return types (!void, !Type) - var declarations at module level - const qualifier on fn bodies - Keyword-style test/invariant/bench blocks (given/when/then) - Dotted function names (Parser.new) - Block comments (/* ... */) - Nested brackets/parens in skip-to-semicolon - Pointer return types (*Type) Results: 27/27 spec files PASS, 15/15 compiler specs PASS Closes #4 Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

…xt (#649) Extends holo_op alphabet in coq/IGLA/RMarker.v with OP_LUT_NPU (TRI-27 ISA 0xE3, Lever #9 LUT-NPU 81-entry hardware port of Microsoft bitnet.cpp lookup table for b1.58 ternary inference). New trios-coq/IGLA/LutNpu.v proves depth-6 alphabet chain via Theorem lut_npu_safe (Forall rtl_uses_star=false), plus 11 spot lemmas (length, head, last, membership, distinctness, extension of W33 TENET chain). Total 16 new Qed accepted by coqc 8.20.1 (W35-G2 >= 10): - RMarker.v: lut_npu_no_star + 3 distinctness lemmas - LutNpu.v: lut_npu_safe theorem + 11 spot lemmas Projection: x1.20 TOPS/W -> 270 TOPS/W on TTIHP27a generic synth (W34 baseline 225). Area cost +0.18 mm^2, power +6 mW. R-SI-1 preserved: OP_LUT_NPU enumerated as rtl_uses_star = false. R7 falsifier W-104-A: BitNet b1.58-3B Trinity-loss sparsity >= 50%. Closes #649 Refs trinity-fpga#120 Anchor: phi^2 + phi^-2 = 3 . DOI 10.5281/zenodo.19227877

…xt (#649) (#651) Extends holo_op alphabet in coq/IGLA/RMarker.v with OP_LUT_NPU (TRI-27 ISA 0xE3, Lever #9 LUT-NPU 81-entry hardware port of Microsoft bitnet.cpp lookup table for b1.58 ternary inference). New trios-coq/IGLA/LutNpu.v proves depth-6 alphabet chain via Theorem lut_npu_safe (Forall rtl_uses_star=false), plus 11 spot lemmas (length, head, last, membership, distinctness, extension of W33 TENET chain). Total 16 new Qed accepted by coqc 8.20.1 (W35-G2 >= 10): - RMarker.v: lut_npu_no_star + 3 distinctness lemmas - LutNpu.v: lut_npu_safe theorem + 11 spot lemmas Projection: x1.20 TOPS/W -> 270 TOPS/W on TTIHP27a generic synth (W34 baseline 225). Area cost +0.18 mm^2, power +6 mW. R-SI-1 preserved: OP_LUT_NPU enumerated as rtl_uses_star = false. R7 falsifier W-104-A: BitNet b1.58-3B Trinity-loss sparsity >= 50%. Closes #649 Refs trinity-fpga#120 Anchor: phi^2 + phi^-2 = 3 . DOI 10.5281/zenodo.19227877 Co-authored-by: Vasilev Dmitrii <admin@t27.ai>

…ernary logic) (Closes #735) (#736) - ring-097-rust: bounded proof trace + K3 ternary logic mirroring specs/ar/proof_trace.t27 byte-for-byte - 823 LOC, 29 tests, all green on Rust 1.83.0 - Cross-kernel anchor #9: cot_phi_identity routes phi^2 + 1/phi^2 = 3 through 6-step proof trace + pow_u64 numeric witness + phi^2/phi^-2 mass-conservation - LOC correction: claimed 624 -> actual 823 (R5-HONEST) Closes #735 Co-authored-by: Trinity Agent <agent@trinity-s3ai.local>

feat(ring-1): lex all 28 specs without errors [SEED-1]

…xt (#649) (#651) Extends holo_op alphabet in coq/IGLA/RMarker.v with OP_LUT_NPU (TRI-27 ISA 0xE3, Lever #9 LUT-NPU 81-entry hardware port of Microsoft bitnet.cpp lookup table for b1.58 ternary inference). New trios-coq/IGLA/LutNpu.v proves depth-6 alphabet chain via Theorem lut_npu_safe (Forall rtl_uses_star=false), plus 11 spot lemmas (length, head, last, membership, distinctness, extension of W33 TENET chain). Total 16 new Qed accepted by coqc 8.20.1 (W35-G2 >= 10): - RMarker.v: lut_npu_no_star + 3 distinctness lemmas - LutNpu.v: lut_npu_safe theorem + 11 spot lemmas Projection: x1.20 TOPS/W -> 270 TOPS/W on TTIHP27a generic synth (W34 baseline 225). Area cost +0.18 mm^2, power +6 mW. R-SI-1 preserved: OP_LUT_NPU enumerated as rtl_uses_star = false. R7 falsifier W-104-A: BitNet b1.58-3B Trinity-loss sparsity >= 50%. Closes #649 Refs trinity-fpga#120 Anchor: phi^2 + phi^-2 = 3 . DOI 10.5281/zenodo.19227877 Co-authored-by: Vasilev Dmitrii <admin@t27.ai>

…ernary logic) (Closes #735) (#736) - ring-097-rust: bounded proof trace + K3 ternary logic mirroring specs/ar/proof_trace.t27 byte-for-byte - 823 LOC, 29 tests, all green on Rust 1.83.0 - Cross-kernel anchor #9: cot_phi_identity routes phi^2 + 1/phi^2 = 3 through 6-step proof trace + pow_u64 numeric witness + phi^2/phi^-2 mass-conservation - LOC correction: claimed 624 -> actual 823 (R5-HONEST) Closes #735 Co-authored-by: Trinity Agent <agent@trinity-s3ai.local>

…xt (#649) (#651) Extends holo_op alphabet in coq/IGLA/RMarker.v with OP_LUT_NPU (TRI-27 ISA 0xE3, Lever #9 LUT-NPU 81-entry hardware port of Microsoft bitnet.cpp lookup table for b1.58 ternary inference). New trios-coq/IGLA/LutNpu.v proves depth-6 alphabet chain via Theorem lut_npu_safe (Forall rtl_uses_star=false), plus 11 spot lemmas (length, head, last, membership, distinctness, extension of W33 TENET chain). Total 16 new Qed accepted by coqc 8.20.1 (W35-G2 >= 10): - RMarker.v: lut_npu_no_star + 3 distinctness lemmas - LutNpu.v: lut_npu_safe theorem + 11 spot lemmas Projection: x1.20 TOPS/W -> 270 TOPS/W on TTIHP27a generic synth (W34 baseline 225). Area cost +0.18 mm^2, power +6 mW. R-SI-1 preserved: OP_LUT_NPU enumerated as rtl_uses_star = false. R7 falsifier W-104-A: BitNet b1.58-3B Trinity-loss sparsity >= 50%. Closes #649 Refs trinity-fpga#120 Anchor: phi^2 + phi^-2 = 3 . DOI 10.5281/zenodo.19227877 Co-authored-by: Vasilev Dmitrii <admin@t27.ai>

…ernary logic) (Closes #735) (#736) - ring-097-rust: bounded proof trace + K3 ternary logic mirroring specs/ar/proof_trace.t27 byte-for-byte - 823 LOC, 29 tests, all green on Rust 1.83.0 - Cross-kernel anchor #9: cot_phi_identity routes phi^2 + 1/phi^2 = 3 through 6-step proof trace + pow_u64 numeric witness + phi^2/phi^-2 mass-conservation - LOC correction: claimed 624 -> actual 823 (R5-HONEST) Closes #735 Co-authored-by: Trinity Agent <agent@trinity-s3ai.local>

…: SGUP-600cell competitor intel - VcdRecorder::record: value: u32 → u64 (#971.1). Fixes silent truncation for signals wider than 32 bits (ternary regfile = 54 bits). - TbConfig::new: add clock_period_ns parameter; compute timeout_ns from max_cycles * clock_period_ns instead of hardcoded *10 (#971.2). Eliminates timeout/SIM PASSED race for non-default clock periods. - COMPETITIVE_POSITIONING.md: add SGUP-600cell (Zenodo:19927449) entry #9 with explicit spectral mass formula m_k = m₀ exp[β_k csc(π α_k / 53)]. - WAVE_LOOP_68_REPORT.md + PLAN.md with metrics and 3 cooperation variants. - Suite 548/548 PASS, cargo test 534/534 PASS. Closes #971 (sub-issues 1 and 2)

…t depth (Refs #1764) (#1904) Variant 1 RTL prototype. Ruled out yosys auto-retiming (57->51, too weak). Built two 2-stage GftSmul prototypes (cut after product; cut mid-RNE), both bit-exact to the combinational core over ~40k random operands, but ltp rose 44->49/50: a register dropped into the generated combinational function defeats yosys cross-function optimization and the RNE/accumulator depth does not split there. Implication: the pipeline must be a spec-level on_clock where the compiler co-optimizes stages (multiply becomes a real pipelined primitive), not a hand-inserted register. Depth reduction is secondary anyway -- the untested lever is mid-cloud resync, and since the microsequencer waits settle >> 1 a latency-1 core needs no sequencer change. Added ruled-out items #8 (auto-retime) and #9 (dropped-in register). Refs #1764 Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>

gHashTag mentioned this pull request Apr 4, 2026

[SEED-1] Ring-1: All immediates — lex all 28 specs without errors #4

Closed

gHashTag merged commit 41516f2 into master Apr 4, 2026
1 check passed

gHashTag deleted the ring/1-lex-all-specs branch April 4, 2026 09:50

gHashTag mentioned this pull request Apr 4, 2026

[SEED-6] Ring-6: test blocks → Zig test blocks #14

Closed

gHashTag mentioned this pull request May 8, 2026

EPIC: FPGA GF16 Matmul Core — Hardware AGI Substrate (Ring 087) #584

Closed

This was referenced May 15, 2026

Wave-35 Lane V — Coq alphabet extension for OP_LUT_NPU (Lever #9) #649

Closed

feat(coq): Wave-35 Lane V LUT-NPU 81-entry bitnet.cpp port alphabet ext (#649) #651

Merged

gHashTag added a commit that referenced this pull request Jun 1, 2026

Merge pull request #9 from gHashTag/ring/1-lex-all-specs

570c046

feat(ring-1): lex all 28 specs without errors [SEED-1]

This was referenced Jun 8, 2026

chore(funding): add maintainer block (Trinity S^3 AI, ORCID 0009-0008-4294-6159) to FUNDING.json #1062

Closed

chore: add maintainer block (Trinity S^3 AI, ORCID 0009-0008-4294-6159) to FUNDING.json #1061

Merged

gHashTag mentioned this pull request Jun 16, 2026

W94 R-RUNTIME: memo cache ignores args, auto-invokes functions on identifier lookup, no cycle detection (CRITICAL) #970

Closed

gHashTag mentioned this pull request Aug 7, 2026

docs: pipeline prototype bit-exact, but dropped-in register won't cut depth -> go spec-level on_clock (Refs #1764) #1904

Merged

Navigation Menu

Repository navigation

feat(ring-1): lex all 28 specs without errors [SEED-1] - #9

feat(ring-1): lex all 28 specs without errors [SEED-1]#9
gHashTag merged 1 commit into
masterfrom
ring/1-lex-all-specs

gHashTag commented Apr 4, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

gHashTag commented Apr 4, 2026

Summary

Changes

Test plan

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants