linux-kernel - Re: [PATCH 2/2] gpio: Make "offset" and "unsigned int", not just "unsigned"

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAD=FV=VtUEgwwB6C4GhTKx3TP37=i4shtwbG41r=wkELcpNmyQ@mail.gmail.com>
Date:   Tue, 28 Apr 2020 18:04:04 -0700
From:   Doug Anderson <dianders@...omium.org>
To:     Joe Perches <joe@...ches.com>
Cc:     LinusW <linus.walleij@...aro.org>,
        Bartosz Golaszewski <bgolaszewski@...libre.com>,
        "open list:GPIO SUBSYSTEM" <linux-gpio@...r.kernel.org>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 2/2] gpio: Make "offset" and "unsigned int", not just "unsigned"

Hi,

On Tue, Apr 28, 2020 at 5:57 PM Joe Perches <joe@...ches.com> wrote:
>
> On Tue, 2020-04-28 at 17:50 -0700, Doug Anderson wrote:
> > Hi,
> >
> > On Tue, Apr 28, 2020 at 5:38 PM Joe Perches <joe@...ches.com> wrote:
> > > On Tue, 2020-04-28 at 17:23 -0700, Douglas Anderson wrote:
> > > > When I copied the function prototypes from the GPIO header file into
> > > > my own driver, checkpatch yelled at me saying that I shouldn't use use
> > > > "unsigned" but instead should say "unsigned int".  Let's make the
> > > > header file use "unsigned int" so others who copy like I did won't get
> > > > yelled at.
> > >
> > > There are a few other unsigned declarations in the file.
> >
> > There are?  I swear I looked for them before I sent my patch and I
> > couldn't find them.  Then I looked again upon seeing your reply and I
> > still can't find them.  My eyes are bad, though.  Maybe you can give
> > me specifics?
>
> $ git grep -P -n '\bunsigned\s+(?!int|long)' include/linux/gpio/driver.h
> include/linux/gpio/driver.h:352:                                                unsigned offset);
> include/linux/gpio/driver.h:354:                                                unsigned offset);
> include/linux/gpio/driver.h:356:                                                unsigned offset);
> include/linux/gpio/driver.h:358:                                                unsigned offset);
> include/linux/gpio/driver.h:360:                                                unsigned offset, int value);
> include/linux/gpio/driver.h:362:                                                unsigned offset);
> include/linux/gpio/driver.h:367:                                                unsigned offset, int value);
> include/linux/gpio/driver.h:372:                                              unsigned offset,
> include/linux/gpio/driver.h:375:                                                unsigned offset);
> include/linux/gpio/driver.h:462:                        unsigned offset);
> include/linux/gpio/driver.h:660:int gpiochip_generic_request(struct gpio_chip *gc, unsigned offset);
> include/linux/gpio/driver.h:661:void gpiochip_generic_free(struct gpio_chip *gc, unsigned offset);
> include/linux/gpio/driver.h:662:int gpiochip_generic_config(struct gpio_chip *gc, unsigned offset,

...riiiiiggght.   ...and now I run your sed script _after_ my patch
and I get no hits.  ...so I'm still confused about what you want me to
do that's not already done in my patch.


> > > Maybe do all of them (and remove the unnecessary externs)?
> >
> > You mean just remove the word "extern" everywhere in this file?  Sure,
> > I can if you want.
>
> Up to the actual maintainers I suppose.
> There are only a few extern function declarations.
> Most do not use extern.

OK, maybe I'll wait for Linux W. or Bartosz to weigh in unless there
is some Linux policy against using "extern" in header files?


> > > trivial reformatting of the function pointer block too
> >
> > Wow, I must be totally out of it.  Maybe it's the gin and tonic I just
> > had.  I don't understand this comment either.  Can you clarify?
>
>         int                             (*foo)(...,
>                                                ...);
>
> might be better with fewer tabs between return type and function pointer
>
>         int             (*foo)(..., ...);

I'll wait for Linux W. or Bartosz to weigh in here, since it feels
more like a style decision.  Happy to add a patch for it, though.

-Doug