Re: [PATCH v2 1/5] HID: magicmouse: register power supply

On Wed, Jul 28, 2021 at 11:35:20AM +0200, Jiri Kosina wrote:

On Fri, 25 Jun 2021, José Expósito wrote:

quoted

quoted

quoted

[...]
v2: Add depends on USB_HID to Kconfig

Hmm, why is this dependency needed in the first place, please? I think 
trying to keep the drivers independent on transport drivers (especially in 
cases like this, where more variants of physical transports actually 
really do exist) is worth trying.

Sorry, that's something I should have explained in the changelog.

Intel's test bot reported compilation errors on the first version of the patch
when USB support wasn't configured:
https://lore.kernel.org/patchwork/patch/1425313/

I was kindly pointed to a similar error and its fix, but, maybe in this case this
is not the right fix?
Maybe there is a macro that I can use to wrap the USB related code in an #ifdef?

It can certainly be wrapped, but looking into the code now, it probably 
wouldn't really bring more clarity. I will apply the series with adding 
the USB_HID dependency for now.

Hi Jiri,

I've been investigating a bit about this issue and I think this might not be the
righ solution for the problem.

John Chen's patch (9de07a4e8d4cb269f9876b2ffa282b5ffd09e05b):
https://lore.kernel.org/lkml/20210327130508.24849-5-johnchen902@gmail.com/ (local)

Already adds battery reporting over bluetooth, so my patch is redundant... And worse
than his, I should add.

I was investigating how to do something similar over USB, but I couldn't finish a patch yet.

So, if you don't mind, I'd prefer not to apply this patchset yet until I figure out
a better solution on v3.

Thanks,
Jose