F0025 err-5¶
Rule: FDCAN one-shot mode retransmits after a lost arbitration.
| Code | F0025 (stable, never reused) |
| Id | err-5 |
| Category | err |
| Source | spirit_level |
| Fix | by hand |
Level per tier¶
| Tier | Level | Reason |
|---|---|---|
| small | off | |
| small-workspace | off | |
| medium | off | |
| big | off | |
| ai | off | |
| embedded | warn |
bedrock warning: FDCAN DAR (one-shot) without retransmit after a lost arbitration on the first two ID bits (ES0491 2.22.3) |
| ui | off | |
| egui | off | |
| units-strict | off | |
| units-paranoid | off |
Why¶
ES0491 2.22.3 ("DAR mode transmission failure due to lost arbitration", H7), status A:
In DAR mode, the transmission may fail due to lost arbitration at the first two identifier bits.
Workaround: "clear the corresponding Tx buffer transmission request bit TRPx of the FDCAN_TXBRP register and set the corresponding cancellation finished bit CFx of the FDCAN_TXBCF register, then restart the transmission."
Example¶
DAR (one-shot transmission, no automatic hardware retransmit) is bedrock_can's one_shot API option on vb135. The workaround itself โ clear TRPx/CFx and restart from software โ is not implemented: a frame lost this way under DAR is simply dropped (tracked as TPL-53 in this repo's FEATURES.md), which is an accepted gap, not a silent one.
How to fix¶
Either implement the documented software retransmit for DAR mode, or, if one-shot semantics are only ever used where an occasional dropped frame at lost arbitration is acceptable, say so explicitly next to the one_shot option and count the drop.
How it is checked¶
Check: src mid ยท chip: H7 โ recognizing "DAR is on and no retransmit path exists" needs the driver's state machine, not a single line.
Source¶
Ported from embedded_bedrock's firmware check catalogue, ERR-5 (book/checks.md, detail in book/checks/err.md). Found on: vb135 PLAT-6.
Configure¶
Allow one finding in the code with // spirit-level: allow(err-5) <reason> (waivers); the code works there too, as in [lints.F0025].