RE: [PATCH v2 2/2] rtw88: add a debugfs entry to enable/disable coex mechanism

[Tony Chuang <yhchuang@realtek.com>]

Kalle Valo <kvalo@codeaurora.org> :

> <yhchuang@realtek.com> writes:
> 
> > From: Yan-Hsuan Chuang <yhchuang@realtek.com>
> >
> > Sometimes we need to stop the coex mechanism to debug, so that we
> > can manually control the device through various outer commands.
> > Hence, add a new debugfs coex_enable to allow us to enable/disable
> > the coex mechanism when driver is running.
> >
> > To disable coex
> > echo 0 > /sys/kernel/debug/ieee80211/phyX/rtw88/coex_enable
> >
> > To enable coex
> > echo 1 > /sys/kernel/debug/ieee80211/phyX/rtw88/coex_enable
> >
> > To check coex dm is enabled or not
> > cat /sys/kernel/debug/ieee80211/phyX/rtw88/coex_enable
> 
> I forgot, did we add a command to nl80211 for managing btcoex? At least
> we have talking about that for years. Please check that first before
> adding a debugfs interface for this.
> 

Yes, I found there was a thread [1] talking about adding a callback to
enable/disable btcoex, but it seems not get applied eventually.

And there's another thread [2] talking about add a btcoex subsystem.
But seems like nobody can implement it cleanly in the host.

I think adding btcoex subsystem could have a lot of pain since each
vendor is using different mechanism controlling the btcoex, and it
usually comes with RF related design which is difficult to write a common
function to deal with all kinds of them.

Yen-Hsuan

[1] https://patchwork.kernel.org/patch/10252135/
[2] https://www.spinics.net/lists/linux-wireless/msg133333.html