40 MHz / 80 MHz coverage left open by #352: 8814AU, 8821AU, Jaguar2 RX, 80 MHz, other channels

未关闭
#554 0 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看

维护者通常 1 天内回复

还没有人认领这个 Issue。

评估

难度
4/5
预计耗时
3-5 天
新手友好度
34/100
Issue 类型
功能
描述清晰度
基本清楚
活跃度
活跃
技术栈
c

调研方向

Start with tests/bw40_pair_onair.sh and docs/vht-on-2g4.md, then review the existing Jaguar2 40 MHz transmit test and the rtw_get_center_ch handling referenced in the issue. The issue calls for measurements across additional chips, receive paths, bandwidths and channels; 80 MHz coverage also needs new cells and rates. Done means recording those measurements with per-leg 20 MHz controls and peer power sweeps.

由索引模型根据 Issue 内容生成。

描述

enhancement

What is left unmeasured after #352

#352 (Jaguar1 40 MHz dead both ways) closed on 2026-11-09: the cause was the HT40 centre-channel tune fixed in #500, verified on an RTL8812AU against an RTL8852CU with tests/bw40_pair_onair.sh (branch bw40-352-verify). That verification covered exactly one die, one channel and one width. These cells are still open:

  • RTL8814AU and RTL8821AU at 40 MHz. Both go through the same rtw_get_center_ch rule as the 8812AU but neither has been measured (noted as unverified in #500 as well). The 8814AU needs the pinned-kernel VM for any vendor-driver control and is judged by its devourer-TX cells only.
  • Jaguar2 DEVOURER_BW=40 receive path. The #352 body reports ad-hoc runs with an 8822BU receiver reading zero at both primary-channel offsets. Jaguar2 TX at 40 MHz is validated by tests/jaguar2_tx_bw40.sh against a kernel sniffer; the devourer receive side at 40 MHz has no record. docs/vht-on-2g4.md no longer claims it is dead, only that it is unmeasured.
  • 80 MHz on Jaguar1 and Jaguar2, both directions, with the same control structure (a 20 MHz control per leg, the wide monitor shown hearing a narrow frame before it is asked to hear a wide one).
  • 2.4 GHz channels other than 6, including the grid-edge fits from #500 (ch1 HT40−, ch13 HT40+ were measured as transmitter only), and at least one 5 GHz channel at 40 MHz on the RX side.

How

tests/bw40_pair_onair.sh already takes --dut, --peer, --channel and --offset, so the first three items are runs, not code — except that the script has no 80 MHz cells yet (DEVOURER_BW=80 / DEVOURER_HOP_BW=80 and the /80 rates would need adding). Keep each leg's 20 MHz control and the peer power sweep (--peer-tx-pwr): on the same-host pair, peer MCS7 into the 8812AU read 0 % at the Kestrel's 20 dBm default and 90 % at 12 or 4 dBm, so a dense-constellation zero must be checked for near-field saturation before it is read as a width defect.

🤖 Generated with Claude Code

主要语言
C
星标
56
派生
29
平均合并
3 小时 43 分钟
30 天内合并 PR
75

环境准备

这个项目没有提供开发容器、Dockerfile 或贡献指南,环境需要你自己搭建:先看它的 README,通用步骤见我们的新手贡献指南。

从这里开始

  1. 先读完整个 Issue,再读项目的贡献指南。
  2. 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
  3. Fork 仓库,在一个分支上完成修改。
  4. 提交 Pull Request,并在描述里引用这个 Issue 编号。

OpenIPC/devourer 的其他 Issue

查看 OpenIPC/devourer 的全部 Issue

相似的 Issue

更多 C Issue

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。