docs: split README 5GHz chip status into UNII-1 vs UNII-2/3#63
Merged
Conversation
The previous "5 GHz" column oversimplified per-chip status. A 2026-06-01
channel sweep across ch1/6/11 (2.4G), ch36/40/44/48 (UNII-1),
ch100 (UNII-2 DFS), and ch149/161 (UNII-3) on all three plugged DUTs
(8812AU, 8821AU, 8814AU) showed a real cutoff at the UNII-1 boundary
for 8821AU:
RTL8812AU devourer TX → kernel RX:
every channel, every band: 4300-4676 hits / 4500 TX ✓
RTL8821AU devourer TX → kernel RX:
ch1/6/11 (2.4G): 4397-4426 hits ✓
ch36/40/44/48 (UNII-1): 0 hits ✗
ch100 (UNII-2 DFS): 0 hits ✗
ch149/161 (UNII-3): 0 hits ✗
RTL8821AU kernel TX → devourer RX (RX-side check):
2.4G: ✓
UNII-1 (ch36/48): 100 hits each ✓
UNII-2/3 (ch100/149/161): 0 hits ✗
RTL8814AU devourer TX → kernel RX:
every channel, every band: 0 hits ✗ (issue #36)
So 8821AU has a sharper boundary than the previous "5 GHz = RX only"
captured: UNII-1 RX works, but UNII-2/3 is completely dark in both
directions. 8814AU TX is broken everywhere as long known.
Table split into a 3-column band breakdown so the chip × band cells
read directly. 8821AU notes also list the five hypotheses tested and
refuted for the 5 GHz TX gate (IQK port from upstream
halrf_iqk_8821a_ce.c, BB-divergence value overrides, missing init
writes like REG_DWBCN1_CTRL_8812, post-channel-set settle delay, TX
power cap) so the next investigator doesn't retry them.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
The previous "5 GHz" column in the hardware-landscape table oversimplified per-chip status. A channel sweep across all three plugged DUTs (8812AU, 8821AU, 8814AU) on 2026-06-01 surfaced a real cutoff at the UNII-1 boundary specifically for 8821AU. This PR splits the "5 GHz" column into "5 GHz UNII-1 (ch36-48)" and "5 GHz UNII-2/3 (ch52+)" so the chip × band cells read directly.
Measurement data
8821AU on TP-Link Archer T2U Plus (
2357:0120), devourer TX → 8812AU kernel RX in VM:8821AU RX side (kernel TX → 8821AU devourer RX):
So 8821AU has a sharper boundary than the previous "5 GHz = RX only" captured: UNII-1 RX works, but UNII-2 / UNII-3 are completely dark in both directions.
For reference, 8812AU works on every channel both bands (4300 – 4676 hits at every channel tested), and 8814AU TX is broken everywhere as long known (issue #36).
What this PR doesn't change
8821AU 5 GHz TX gate investigation is still open. Five hypotheses tested and refuted to date (IQK port from upstream
halrf_iqk_8821a_ce.c, BB-divergence value overrides, missing init writes likeREG_DWBCN1_CTRL_8812, post-channel-set settle delay, TX power cap). The 8821 cell in the updated table now lists these so the next investigator doesn't retry them. See issue #59.Test plan
🤖 Generated with Claude Code