Skip to content

move the chip pin to v1.3.0 - #19

Merged
A13xB0 merged 1 commit into
mainfrom
chip-pin-v1.3.0
Sep 3, 2026
Merged

A13xB0 merged 1 commit into
mainfrom
chip-pin-v1.3.0

Conversation

@A13xB0

@A13xB0 A13xB0 commented Sep 3, 2026

Copy link
Copy Markdown
Collaborator

A submodule is a commit, not a branch, so four commits landed in
MeshBench/virtual-sx1262 and
reached nothing here. That is the whole hazard CONSUMERS.md and the
consumers CI job were written for, and this is the first time the rule has
caught anything: the pin lagged from the moment the byte-path ABI merged, and
nothing here failed or complained.

What comes with it

  • The byte-at-a-time SPI path (ABI 1.3). QEMU and Renode drive it because
    their controllers clock one byte and want the answering byte back before the
    next; a native node has the whole command in a buffer and does not use it.
  • The test file split. test_model.cpp had reached 537 lines against that
    repository's own 500-line limit, so its CI had been red since before the
    extraction.
  • The release matrix fix, without which v1.3.0 could not be cut at all:
    GitHub retired the macos-13 runner, so that row queued for ever and took the
    publish job with it.

Checked

simple_repeater builds for linux-amd64 against this. A native node compiles
the model in, which is what keeps it the same chip the emulators load, from the
same commit.

🤖 Generated with Claude Code

https://claude.ai/code/session_01Q9HbD44EKWWTRYgxbFGxf6

A submodule is a commit, not a branch, so four commits landed in
MeshBench/virtual-sx1262 and reached nothing here. That is the whole hazard
CONSUMERS.md and the consumers CI job were written for, and this is the first
time the rule has caught anything: the pin lagged from the moment the byte-path
ABI merged.

What comes with it: the byte-at-a-time SPI path (ABI 1.3), which is what the
emulators drive and which a native node does not use; the test file split, which
had put that repository's CI red against its own 500-line limit; and the release
matrix fix that let v1.3.0 be cut at all.

A native node compiles the model in, which is what keeps it the same chip the
emulators load from the same commit. simple_repeater builds for linux-amd64
against this.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Q9HbD44EKWWTRYgxbFGxf6
@A13xB0
A13xB0 merged commit 8b3e8ec into main Sep 3, 2026
1 check passed
@A13xB0
A13xB0 deleted the chip-pin-v1.3.0 branch September 3, 2026 23:56
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.

2 participants