Miaoqing Pan <quic_miaoqing@xxxxxxxxxxx> writes: >>>>>> Understand your concern, automatic adaptation is always the best >>>>>> choice. But >>>>>> it may not work for MSM boards, the PCIe card (non m.2) is >>>>>> customized, which >>>>>> has special PMU control. User can't swap cards. And that's why power >>>>>> sequencing module was introduced. >>>>> >>>>> I know. Still, it's better to have less unnecessary data there for >>>>> autodiscoverable devices. >>>> >>> >>> We discussed internally, we have no other choice to enable NFA765 for non >>> X86 boards. Could you please approve this 'DT' approach ? >> If you can't use subdevice approach for some reason, then we have no >> other choice that I can imagine. >> > > A new patch was submitted: > https://lore.kernel.org/linux-wireless/20241031000541.3331606-1-quic_miaoqing@xxxxxxxxxxx/. > This patch will add QCA6698AQ support, which follows the approach done > in commit 5dc9d1a55e95 ("wifi: ath11k: add support for QCA2066"), > enumerates the subversion number to identify the specific card. > > But there is still a problem enabling NFA765 m.2 card for IoT > platforms, which requires ath11k to support board-specific firmware > overrides. So there are multiple different hardware you want to support? This is very confusing and the commit message does not really tell anything about those. Can you list _all_ the hardware you want to support and what firmware it needs? -- https://patchwork.kernel.org/project/linux-wireless/list/ https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
