Measure & wire USB-Kestrel TSF-write / TBTT-follow so a USB 8852CU can run a cell node (M6 moving-UE)
維護者通常 1 天內回覆
還沒有人認領這個 Issue。
評估
- 難度
- 5/5
- 預估耗時
- 一週以上
- 新手友好度
- 12/100
- Issue 類型
- 功能
- 描述清晰度
- 基本清楚
- 活躍度
- 活躍
- 技術堆疊
- cpp
- 領域
- networking
研究方向
Start in src/kestrel/RtlKestrelDevice.cpp where tsf_write_ok and tbtt_follows_tsf are forced false for USB, and read the PCIe path that sets them from measurement. The USB analogue to build on is tests/kestrel_tbtt_probe.cpp (PCIe/8852C only) plus the cellcaps probe in examples/cellcaps, which produced the caps dump in the issue. Done means a measured TSF write and TBTT-follow on a USB 8852CU, the two caps gated on that measurement, and a cell UE landing its uplink in slot. This needs the physical radio on the bench, so it is not a code-only start.
由索引模型根據 Issue 內容生成。
描述
Part of the scheduled-RAN epic #254; unblocks the portable moving-UE node in #563 (Option B) and the M6 follow-ups (#566).
What
A USB Kestrel (RTL8852CU/8832CU) cannot run a cell node today, because the cell caps gate (examples/cell caps_ok) requires tsf_write_ok ∧ tbtt_follows_tsf ∧ beacon_backoff_ctl ∧ hw_beacon_txtsf, and over USB the first two are false by policy, not by measurement:
// src/kestrel/RtlKestrelDevice.cpp
c.tsf_write_ok = !_device.is_usb(); // "USB is unmeasured, so it stays refused there"
c.tbtt_follows_tsf = !_device.is_usb();
This was confirmed on hardware (an RTL8852CU on z890p, read with the cellcaps probe):
{"radio":"usb 0bda:c85a","chip":"RTL8852C","variant":"C8852C","transport":"usb",
"tsf_write_ok":0,"tbtt_follows_tsf":0,"beacon_backoff_ctl":1,"hw_beacon_txtsf":1,
"ack_responder_ok":1,"hw_rx_timestamp":1,"hw_injected_mgmt_txtsf":1,"cell_node_ok":0}
So it is the same 8852C die as the fleet's 8852CE, correctly identified, with the ACK responder, HW beacon TX-TSF and RX timestamp all present — only the TSF-write / TBTT-follow pair is withheld because it has never been validated over the USB bus.
Why it matters
The M6 moving-UE bench (#563, Option B) needs a portable, battery-friendly UE node. A USB 8852CU is the obvious candidate — but a UE still needs caps_ok (its uplink is a TSF-steered HW beacon). Until the USB TSF path is measured and wired, no USB radio can be a cell UE, and the moving-UE walk has no portable node.
To do
- Measure, over USB, on the bench (8852CU/8832CU):
WriteTsfactually moves the TSF by the requested step — a read-add-write against the free-running counter, as the PCIe RTL8852CE bench did (the basis for the PCIetsf_write_ok).- TBTT follows the TSF: a
WriteTsfon an active HW beacon moves its TBTT with it — the USB analogue oftests/kestrel_tbtt_probe.cpp(currently PCIe + 8852C only), witnessed on a second radio.
- Wire it: set
tsf_write_ok/tbtt_follows_tsftrue for USB Kestrel gated on the measurement (by variant / measured flag), socaps_okpasses only where it has been shown to hold. - Rig check: run a cell UE on a USB 8852CU against a known-good DU (over-the-air sync, no PTP needed) and confirm its uplink lands in slot, as the PCIe UEs do.
Notes
beacon_backoff_ctl,hw_beacon_txtsf,ack_responder_ok,hw_rx_timestampare already true over USB, so only the TSF-write/TBTT pair is in question.- The
cellcapsprobe used above (examples/cellcaps, reads the cell-gate caps for any USB/PCIe radio) can be PR'd alongside if useful. - DU use over USB additionally needs PTP, which is a separate matter; this issue is scoped to the UE caps.
🤖 Generated with Claude Code
- 主要語言
- C
- 星號
- 56
- 分支
- 29
- 平均合併
- 3 小時 43 分鐘
- 30 天內合併 PR
- 75
環境準備
這個專案沒有提供開發容器、Dockerfile 或貢獻指南,環境需要你自己搭建:先看它的 README,通用步驟見我們的新手貢獻指南。
從這裡開始
- 先讀完整個 Issue,再讀專案的貢獻指南。
- 在 Issue 下留言說明你要接手 —— 這能避免兩個人做同樣的事。
- Fork 儲存庫,在一個分支上完成修改。
- 送出 Pull Request,並在描述裡引用這個 Issue 編號。
OpenIPC/devourer 的其他 Issue
-
難度 2/5 1-3 小時 新手友好度 82/100
維護者通常 1 天內回覆
-
Follow-ups left open by the 8822B fast bandwidth toggle fix (#584)可能已有人在做 @josephnef 今天認領。 未關閉enhancement
難度 4/5 3-5 天 新手友好度 42/100
維護者通常 1 天內回覆
-
enhancement
難度 5/5 一週以上 新手友好度 12/100
維護者通常 1 天內回覆
-
enhancement
難度 5/5 一週以上 新手友好度 8/100
維護者通常 1 天內回覆
-
M6 test realism — a moving-UE handover bench (conducted path-loss matrix, then an over-the-air walk)未關閉enhancement
難度 5/5 一週以上 新手友好度 25/100
OpenIPC/devourer#563 · 1 則留言 ·
維護者通常 1 天內回覆
相似的 Issue
-
難度 2/5 1-3 小時 新手友好度 72/100
維護者通常 1 天內回覆
-
Needs-Triage
難度 2/5 1-3 小時 新手友好度 66/100
維護者通常 2 天內回覆
-
issue: bug report
難度 2/5 1-3 小時 新手友好度 82/100
維護者通常 2 天內回覆
-
cross-platform fix-now
難度 2/5 1-3 小時 新手友好度 76/100
維護者通常 1 天內回覆
-
feature:index kind:crash status:confirmed triaged
難度 2/5 1-3 小時 新手友好度 77/100
維護者通常 1 天內回覆