Skip to content

Pitfall: ApolloControl IDESPEED 3 can corrupt writes to a CF card - #61

Open
WDrijver wants to merge 1 commit into
mainfrom
pitfall/ide-speed-cf-corruption
Open

WDrijver wants to merge 1 commit into
mainfrom
pitfall/ide-speed-cf-corruption

Conversation

@WDrijver

Copy link
Copy Markdown
Contributor

What

A new hardware pitfall, hardware/pitfalls/ide-speed-cf-write-corruption.md (the first file in hardware/pitfalls/).

ApolloControl ID=3 (IDESPEED "Fastest") corrupted writes to a CF card on the V4 IDE port: an SFS block written at ID=3 had a wrong checksum, and it stayed wrong on every reread at ID=2. With ID=0, 1 and 2, a 16 MB file was written and checked by SHA-256 on the V4 without errors, at about the same speed (write 2.7–2.9 s). So ID=3 gained nothing on this card.

The advice is to measure each card (write, hash, reread at a lower level) before putting an IDE speed into the Startup-Sequence. Only one card was tested, and the file says so.

Found during

The ApolloOS R10 port, moving its work volume to a CF card with SFS (V4 UniCorn, Core 12856, SanDisk SDCFXS 128 GB).

Knowledge base files relied on

  • guides/testing-on-v4-hardware.md (ApolloControl, test setup)
  • CONTRIBUTING.md (folder, frontmatter, evidence table)

Local checks: check_frontmatter, check_links and check_saga_addresses pass.

🤖 Generated with Claude Code

Measured on a V4 UniCorn (Core 12856) with a SanDisk 128 GB CF card:
ID=0..2 write and read correctly at about the same speed, ID=3 writes
corrupted blocks (SFS checksum error, same on every reread).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@WDrijver
WDrijver requested a review from TuKo1982 as a code owner November 17, 2026 16:15
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.

1 participant