theater/cabletest
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Re-transcribe BCM84891L MDIO command and SFP register excerpts in full (all 45 commands, complete serial-ID table); scrub doc-provenance attribution and label all our interjections as field notes

Browse Source
This commit is contained in:
flamingcow
2026-08-12 21:02:30 -07:00
parent 5588ebec6f
commit 9cb82ca3fa
3 changed files with 744 additions and 237 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/modules/fs/README.md
+4 -2
View File
@@ -52,7 +52,7 @@ Several documented DATA1 returns on this ODM firmware are untrustworthy: die-tem
## ECD — recovered from the OpenBCM SDK, proven on hardware
The ECD register mechanism is absent from the handler catalog and from FS's docs, but the OpenBCM SDK's copper-XGPHY driver (`sdk-6.5.27/src/soc/phy/phy8481.c` `phy_8481_cable_diag` + `phy8481.h`) carries it for the 8483x/8485x/8488x family — and the same SDK drives the identical command-handler registers (1E.0x4005/0x4037/0x4038–3C) as the BCM84891L datasheet, confirming the shared map. Validated on the FS:
The ECD register mechanism is absent from the handler catalog and the datasheet excerpts here, but the OpenBCM SDK's copper-XGPHY driver (`sdk-6.5.27/src/soc/phy/phy8481.c` `phy_8481_cable_diag` + `phy8481.h`) carries it for the 8483x/8485x/8488x family — and the same SDK drives the identical command-handler registers (1E.0x4005/0x4037/0x4038–3C) as the BCM84891L datasheet, confirming the shared map. Validated on the FS:
| Register | Role |
|---|---|
@@ -67,7 +67,9 @@ Observed on a plugged, linked, healthy cable: verdicts OK×4 and per-pair length
- Family constraints from the SDK: port must be enabled; unsupported at forced 100M.
- `bcm_ecd_probe.py` in phydiag-work implements the recipe.
## Remaining asks to FS
## Missing datasheet chapters
Chapters the datasheet TOC lists but the excerpts here omit — to source elsewhere:
1. The **ECD chapter** — now for confirming bit meanings rather than unblocking.
2. The **1588 operation chapter** (in-PHY timestamping would measure path delay at the MDI, removing PHY-pipeline latency from the length equation; [../../open-questions.md](../../open-questions.md) §2).