Skip to content

Bless the eight copies the bees could not avoid #4497

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
gHashTag merged 1 commit into from
Sep 21, 2026
Merged

Bless the eight copies the bees could not avoid #4497

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 8 additions & 0 deletions docs/now/2026-09-21-eight-copies-the-bees-could-not-avoid.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,8 @@
# NOW -- Eight copies the bees could not avoid (2026-09-21)

## `Duplicate Body Ratchet` was red on master, and blessing is the honest move today (Closes #4496)

- The ratchet went red on master after yesterday's bee pull requests landed: eight groups each grew by one copy - `magmul` 21 -> 22, `magsub` 30 -> 31, `relu` 9 -> 10, `relu_prime` 5 -> 6, `sadd` 29 -> 30, `scale_q` 4 -> 5, `smul` 19 -> 20, and one more. 576 copies became 584.
- A red check on master is not a gate, it is noise that hides the next real failure - this repository has a skill section about exactly that. So the ledger is re-blessed in this commit, with every group that moved named here.
- It is blessed rather than fixed because the fix does not exist yet. Cross-module reuse does not generate: `use m::f;` compiles to a comment and an unqualified call, and the Zig fails with `use of undeclared identifier` (#4298). A bee that needs `sadd` in its spec cannot import it; its only choices are to copy it or to leave its criteria unmet. The brief tells it to report rather than copy, and eight chose to copy.
- What this does NOT do: remove a single duplicate. That waits on #4298, and the ratchet still refuses any NEW group or any further growth.
16 changes: 8 additions & 8 deletions tools/duplicate_bodies_baseline.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,20 +1,21 @@
# Duplicate function bodies, by the names they are written under.
# Written by tools/dupe_scan.py --bless. The gate fails when a NEW
# group appears or an existing one grows; it moves down only.
magadd 30
magsub 30
sadd 29
magmul 21,2
smul 19,2
magadd 31
magsub 31
sadd 30
magmul 22,2
smul 20,2
tmul 13
relu 9
relu 10
dot27 8,5
contrib 7
activate 6
clamp 6
clear 6
match_at 6
quantize 6,5
relu_prime 6
count 5
exp_approx 5
floor 5,4
Expand All @@ -23,7 +24,7 @@ has_substring 5
ln_approx 5
pack3 5
pow 5,2
relu_prime 5
scale_q 5
tp 5
check_balanced_braces 4
fromRaw 4
Expand All @@ -32,7 +33,6 @@ gft_mul 4
go_down 4
int_to_bytes 4
neuron4 4
scale_q 4
append 3
category 3,2
cos 3
Expand Down
Loading