Re: [PATCH 1/1] wifi: mt76: mt7996: fix mixed FEM check of mt7996 chipset

Valera Kozlov <[email protected]>
Newsgroups org.kernel.vger.linux-wireless,org.infradead.lists.linux-arm-kernel,org.infradead.lists.linux-mediatek,org.kernel.vger.linux-kernel
Message-ID <[email protected]>
Device: ASUS ZenWiFi BT8 (RT-BE14000), MediaTek MT7988D SoC (tri-band,
dual PCIe hif/hif2). Confirmed empirically via extended TX PER testing
that adie_idx=1 is correct for this board (matches Chukun's patch),
consistent with a mixed-FEM design.

Readouts from the ADIE_ID registers:

root@bt8:~# echo 0x0f00002c >/sys/kernel/debug/ieee80211/phy0/mt76/regidx
root@bt8:~# grep . /sys/kernel/debug/ieee80211/phy0/mt76/rf_regval
0x79768a20
root@bt8:~# echo 0x1f00002c >/sys/kernel/debug/ieee80211/phy0/mt76/regidx
root@bt8:~# grep . /sys/kernel/debug/ieee80211/phy0/mt76/rf_regval
0x79778a10
root@bt8:~# echo 0x2f00002c >/sys/kernel/debug/ieee80211/phy0/mt76/regidx
root@bt8:~# grep . /sys/kernel/debug/ieee80211/phy0/mt76/rf_regval
0x79778a10

Note that band1/band2 (0x79778a10) match the EE4600 signature exactly
(needs adie_idx=1), but band0 (0x79768a20) differs from both Zyxel
devices reported so far (both show 0x79758a02 regardless of which
adie_idx they need). This is a third, distinct band0 signature -
suggests band0 alone may not reliably disambiguate the two cases,
since devices needing different adie_idx values can share it.

Happy to test further register dumps or a proposed fix on this
hardware if useful.
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.