Re: [PATCH v1] Input: Drop unused assignments from pnp_device_id arrays

[Uwe Kleine-König (The Capable Hub)]

Hello Dmitry,

On Tue, Jun 09, 2026 at 11:12:49PM -0700, Dmitry Torokhov wrote:
> On Tue, Jun 09, 2026 at 04:53:25PM +0200, Uwe Kleine-König (The Capable Hub) wrote:
> > Explicitly assigning .driver_data in drivers that don't use this member
> > is silly and a bit irritating. Drop these. Also simplify the list
> > terminator entry to be just empty to match what most other device_id
> > tables do.
> > 
> > There is no changed semantic, not even a change in the compiled result.
> > 
> > Signed-off-by: Uwe Kleine-König (The Capable Hub) <u.kleine-koenig@xxxxxxxxxxxx>
> > ---
> >  drivers/input/gameport/ns558.c        | 46 +++++++++++-----------
> >  drivers/input/serio/i8042-acpipnpio.h | 56 +++++++++++++--------------
> >  2 files changed, 51 insertions(+), 51 deletions(-)
> > 
> > diff --git a/drivers/input/gameport/ns558.c b/drivers/input/gameport/ns558.c
> > index fdece6ec1df3..f70a96c4f1fd 100644
> > --- a/drivers/input/gameport/ns558.c
> > +++ b/drivers/input/gameport/ns558.c
> > @@ -148,29 +148,29 @@ static int ns558_isa_probe(int io)
> >  #ifdef CONFIG_PNP
> >  
> >  static const struct pnp_device_id pnp_devids[] = {
> > -   [...]
> > -	{ .id = "", },
> > +   [...]
> > +	{ }
> 
> This goes BOOOM! You have to keep empty .id string as terminator.

How so? Given that my patch doesn't modify the resulting ns558.o I doubt
that. If .id was a char *, I'd agree, but it's a char[], so there should
be no difference (and the compiler agrees).

Best regards
Uwe

Attachment: signature.asc
Description: PGP signature