MINT-ZUSE says "nothing changed" after it has rewritten and zeroed the stick #168

New Issue

2026-11-11T21:31:30Z

rajames commented

2026-11-11 21:31:30 +00:00

Found

kernel/src/v4/sk_v4_mint.c: failed() (line 68) prints MINT-ZUSE: failed -- nothing changed. It is called after gpt_write_one() (line 112) has already rewritten the stick's GPT and after the identity partition's first 32 Forth blocks are zeroed (lines 123-124) -- from the GPT read-back, the zeroing, the mint (127), the read-back check (131) and the marker write (138). On those paths the stick has been erased; only the refusals before line 112 (no stick, under 64 MiB, no YES, 512-byte sectors) truly change nothing. The mint_fail test (FABRIC-3 XLVIII.17) covers a stick whose writes all fail, where the message is true.

Done when

A test where the GPT write succeeds and a later step fails, seen to fail on the message
The message on each path says what was written (e.g. "the stick was erased; Zuse not minted")

🤖 Filed with Claude Code -- the code review of 2026-11-11, at 4d42d25c (Captain Bob, 2026-11-11: "File the review findings as issues"). Read, not run: each is to be confirmed by a test written first and seen to fail.

### Found `kernel/src/v4/sk_v4_mint.c`: `failed()` (line 68) prints `MINT-ZUSE: failed -- nothing changed`. It is called after `gpt_write_one()` (line 112) has already rewritten the stick's GPT and after the identity partition's first 32 Forth blocks are zeroed (lines 123-124) -- from the GPT read-back, the zeroing, the mint (127), the read-back check (131) and the marker write (138). On those paths the stick has been erased; only the refusals before line 112 (no stick, under 64 MiB, no YES, 512-byte sectors) truly change nothing. The `mint_fail` test (FABRIC-3 XLVIII.17) covers a stick whose writes *all* fail, where the message is true. ### Done when - [ ] A test where the GPT write succeeds and a later step fails, seen to fail on the message - [ ] The message on each path says what was written (e.g. "the stick was erased; Zuse not minted") 🤖 Filed with [Claude Code](https://claude.com/claude-code) -- the code review of 2026-11-11, at `4d42d25c` (Captain Bob, 2026-11-11: "File the review findings as issues"). Read, not run: each is to be confirmed by a test written first and seen to fail.

rajames added the area/kernel

labels 2026-11-11 21:31:31 +00:00

Sign in to join this conversation.