[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <2025081746-cyclist-stegosaur-6895@gregkh>
Date: Sun, 17 Aug 2025 11:42:03 +0200
From: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To: Carlos Llamas <cmllamas@...gle.com>
Cc: Salvatore Bonaccorso <carnil@...ian.org>,
Arve Hjønnevåg <arve@...roid.com>,
Todd Kjos <tkjos@...roid.com>, Martijn Coenen <maco@...roid.com>,
Joel Fernandes <joelagnelf@...dia.com>,
Christian Brauner <christian@...uner.io>,
Suren Baghdasaryan <surenb@...gle.com>,
Ben Hutchings <benh@...ian.org>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] binder: Add missing module description
On Mon, Aug 11, 2025 at 06:44:49PM +0000, Carlos Llamas wrote:
> On Mon, Aug 11, 2025 at 08:01:05PM +0200, Salvatore Bonaccorso wrote:
> > Hi,
> >
> > On Mon, Aug 11, 2025 at 05:36:20PM +0000, Carlos Llamas wrote:
> > > On Sat, Aug 09, 2025 at 09:30:18AM +0200, Salvatore Bonaccorso wrote:
> > > > During build modpost issues a warning:
> > > >
> > > > # MODPOST Module.symvers
> > > > ./scripts/mod/modpost -M -m -b -o Module.symvers -n -T modules.order vmlinux.o
> > > > WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/android/binder_linux.o
> > > >
> > > > Fix this by adding the missing module description.
> > >
> > > I'm not against having a description but binder is not tristate. Judging
> > > from the 'binder_linux.o' this seems to be some OOT version? Otherwise,
> > > I'm not sure you would be able to reproduce the warn.
> >
> > This is entirely my fault. We saw the warning while building in
> > Debian, but missed that we apply the following patch:
> > https://salsa.debian.org/kernel-team/linux/-/blob/debian/latest/debian/patches/debian/android-enable-building-binder-as-module.patch?ref_type=heads
> > which allows to build binder as module.
> >
> > I got confused by the fact that we have
> > MODULE_LICENSE("GPL v2");
> > in the upstream variant, but no description.
> >
> > So you want to still have the description added or drop this proposed
> > patch completely instead?
>
> I think it is best to drop all MODULE_* macros and avoid further
> confusion. Including the current MODULE_LICENSE(), this is misleading.
> I'll send a patch fixing this.
Please do, that should make it more obvious that this can't be a module.
thanks,
greg k-h
Powered by blists - more mailing lists