Hi Sean,

On Mon, Mar 24, 2026 Sean Wang wrote:

I don't think this is correct for mt7925, and it will cause a
regression there. Was this tested on actual mt7925 hardware?

You are right. I do not have mt7925 hardware to verify against.
In v3, is_320mhz_supported() now checks is_mt7927() only, so
mt7925 behavior is unchanged.

I don't think this should be copied from mt7996 as-is for mt7927. I'd
suggest dropping the eht_cap_elem->phy_cap_info[7] change and keeping
it conservative for now.

Agreed. The phy_cap_info[7] additions (NON_OFDMA_UL_MU_MIMO_320MHZ
and MU_BEAMFORMER_320MHZ) are dropped in v3. Only the MCS/NSS maps
and per-BW beamformee SS/sounding dimensions for 320MHz remain.

Both changes are in v3, sent today.

Best,
Javier