Re: [PATCH] staging: greybus: remove driver depending on nonexistent config option

[Agatha Isabelle Moreira]

On Tue, Jun 09, 2026 at 06:14:41PM -0700, Ethan Nelson-Moore wrote:
> Hi, Agatha,

Hi, Ethan!

> 
> On Tue, Jun 9, 2026 at 4:56 PM Agatha Isabelle Moreira <code@xxxxxxxxxx> wrote:
> > On Tue, Jun 09, 2026 at 01:26:58PM -0700, Ethan Nelson-Moore wrote:
> > > The Greybus Arche Platform driver depends on the config option
> > > USB_HSIC_USB3613, which has never existed in mainline Linux. Therefore,
> >
> > Actually it dosn't. The `depends on` line says:
> >
> >      depends on USB_HSIC_USB3613 || COMPILE_TEST
> >
> > That's a logical OR operation, it depends on USB_HSIC_USB3613 OR
> > COMPILE_TEST.
> 
> The function of COMPILE_TEST is to allow drivers to be compiled in
> environments where they do not work, to enable better code coverage
> and easier testing of tree-wide changes without cross compilers. It

Yes, and that's been also discussed in the thread I linked on my
previous message as well. My point is just on commit message accuracy.

> should never be enabled for production use cases. So yes, technically,

Not the main point here, but I'd also note that no staging driver is
intended for production use cases.

> it does depend on either option, but only on one "real" option.
> Therefore, I don't think there is anything wrong with the commit
> message.

While I do understand that this driver only compiles and cannot be
loaded, I think this piece of information isn't accurately described in
the commit message, especially when considering it's been argued about
that in previous patches.

Sincerely, 
Agatha Isabelle Moreira