Skip to content

compiler.rs: delete the tests calling the seven items #10896 removed (Closes #10957) - #10958

Merged
gHashTag merged 3 commits into
masterfrom
claude/fix-compiler-dead-code
Oct 19, 2026
Merged

gHashTag merged 3 commits into
masterfrom
claude/fix-compiler-dead-code

Conversation

@gHashTag

@gHashTag gHashTag commented Oct 19, 2026 •

Copy link
Copy Markdown
Owner

Closes #10957

Two compiler.rs fixes for what #10896 left on master. Neither is built locally (no builds on the Mac); CI compiles and checks them.

  1. feat(ml): thread primitive + train_hslm windows on N threads, bit-identical at any count (#10876) #10896 removed seven unused items to keep the foreign-line budget net-negative, but eight test lines still called five of them, so the t27c unit tests did not compile (build-and-test and test-ratchet red since). Restoring the items is refused by the foreign-budget gate, so this deletes the tests that call them.
  2. feat(ml): thread primitive + train_hslm windows on N threads, bit-identical at any count (#10876) #10896 appends the thread runtime to any Zig output that calls thread_spawn, thread_join or thread_cores. compiler/runtime/runtime.t27 defines and calls its own thread_spawn, so its gen_hash_zig changed and the triruntime and runtime_triruntime seals drifted (the coverage check passes on every base before feat(ml): thread primitive + train_hslm windows on N threads, bit-identical at any count (#10876) #10896 and fails on 356e7ce). The detection is now guarded by declared_fns, as the call rewrite already is. One line changed, none added. The third red seal, t27f_T27fFrontend, is another PR's.

The compiler.rs seal (bootstrap/stage0/FROZEN_HASH) moves with the edits. Rust lines only go down or stay level.

🤖 Generated with Claude Code

…d, move the seal (Closes #10957)

#10896 dropped Node::with_child, Parser::check_peek, icarus_lowerability_reason,
align_address, reset_end_cycle, path_delay and worst_path_delay to keep the
foreign-line budget net-negative, but eight test lines still called five of them,
so the unit tests no longer compiled. Delete those tests; nothing else changes.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-10-19 08:54:03 UTC

Summary

Status Count
Total Open PRs 50
PRs with Failing Checks 46
PRs with All Checks Green 4
READY 1
FAILING 46
PENDING 0
NO CHECKS YET 0

These columns do not partition: 1 + 46 + 0 + 0 = 47, and there are 50 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=8c44c8a1ee52 != 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).

gHashTag and others added 2 commits October 19, 2026 15:59
…_cores does not get the thread runtime appended (Closes #10957)

#10896 appended the thread runtime to any Zig output that calls thread_spawn,
thread_join or thread_cores, so compiler/runtime/runtime.t27 (which defines and
calls its own thread_spawn) changed its gen_hash_zig and the triruntime and
runtime_triruntime seals drifted (coverage red). Guard the detection with
declared_fns, as the call rewrite already is. One line changed, none added.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-10-19 09:11:12 UTC

Summary

Status Count
Total Open PRs 50
PRs with Failing Checks 47
PRs with All Checks Green 3
READY 2
FAILING 47
PENDING 0
NO CHECKS YET 0

These columns do not partition: 2 + 47 + 0 + 0 = 49, and there are 50 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=2d301cdb4a58 != 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).

@gHashTag
gHashTag enabled auto-merge (squash) October 19, 2026 09:21
@gHashTag
gHashTag merged commit 027a0e2 into master Oct 19, 2026
40 of 48 checks passed
gHashTag pushed a commit to dmitrii-f-t27/t27 that referenced this pull request Oct 19, 2026
…ged compiler.rs (Closes gHashTag#5649, Closes gHashTag#5650, Closes gHashTag#5651)

master's 027a0e2 (gHashTag#10958, dead tests removed from compiler.rs) moved compiler.rs, so FROZEN_HASH conflicted again; compiler.rs merged clean and FROZEN_HASH is t27c frozen-digest of the merged file. master changed no seals in these 2 commits.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
gHashTag pushed a commit to dmitrii-f-t27/t27 that referenced this pull request Oct 19, 2026
…ee more dead compiler.rs methods and their tests (Closes gHashTag#5649, Closes gHashTag#5650, Closes gHashTag#5651)

own-language's budget (owner rule 2026-10-09) failed the merged branch: compiler.rs was +25/-11 because master's gHashTag#10958 had already removed align_address, the dead function this PR used as its offset. rustc reports these never used outside tests: ArcKind::from_i8, RouteEstimate::passed (with test_route_passed and test_route_failed) and CtsReport::passed (with test_cts_report_passed). Removed with the tests that only exercise them. compiler.rs is now +25/-72 against master (net -47); check_budget passes. t27c builds from this tree; FROZEN_HASH is its frozen-digest. The test target cannot be compiled in my sandbox (a dev-dependency crate is not fetchable), so CI's build-and-test is the check for the removed tests.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

compiler.rs: delete the tests calling the seven items #10896 removed (tests do not compile)

1 participant