linux-kernel - Re: [PATCH] staging: greybus: fix gb_loopback_stats

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]

Date:   Fri, 14 May 2021 20:42:16 +0530
From:   Shreyansh Chouhan <chouhan.shreyansh630@...il.com>
To:     Greg KH <gregkh@...uxfoundation.org>
Cc:     pure.logic@...us-software.ie, johan@...nel.org, elder@...nel.org,
        greybus-dev@...ts.linaro.org, linux-staging@...ts.linux.dev,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] staging: greybus: fix gb_loopback_stats_attrs definition

On Fri, May 14, 2021 at 04:30:23PM +0200, Greg KH wrote:
> On Fri, May 14, 2021 at 07:53:57PM +0530, Shreyansh Chouhan wrote:
> > On Fri, May 14, 2021 at 04:05:32PM +0200, Greg KH wrote:
> > > On Fri, May 14, 2021 at 07:18:38PM +0530, Shreyansh Chouhan wrote:
> > > > On Fri, May 14, 2021 at 03:36:25PM +0200, Greg KH wrote:
> > > > > On Fri, May 14, 2021 at 07:00:39PM +0530, Shreyansh Chouhan wrote:
> > > > > > The gb_loopback_stats_attrs macro, (defined in loopback.c,) is a
> > > > > > multiline macro whose statements were not enclosed in a do while
> > > > > > loop.
> > > > > >
> > > > > > This patch adds a do while loop around the statements of the said
> > > > > > macro.
> > > > > >
> > > > > > Signed-off-by: Shreyansh Chouhan <chouhan.shreyansh630@...il.com>
> > > > > > ---
> > > > > >  drivers/staging/greybus/loopback.c | 10 ++++++----
> > > > > >  1 file changed, 6 insertions(+), 4 deletions(-)
> > > > > >
> > > > > > diff --git a/drivers/staging/greybus/loopback.c b/drivers/staging/greybus/loopback.c
> > > > > > index 2471448ba42a..c88ef3e894fa 100644
> > > > > > --- a/drivers/staging/greybus/loopback.c
> > > > > > +++ b/drivers/staging/greybus/loopback.c
> > > > > > @@ -162,10 +162,12 @@ static ssize_t name##_avg_show(struct device *dev,		\
> > > > > >  }									\
> > > > > >  static DEVICE_ATTR_RO(name##_avg)
> > > > > >
> > > > > > -#define gb_loopback_stats_attrs(field)				\
> > > > > > -	gb_loopback_ro_stats_attr(field, min, u);		\
> > > > > > -	gb_loopback_ro_stats_attr(field, max, u);		\
> > > > > > -	gb_loopback_ro_avg_attr(field)
> > > > > > +#define gb_loopback_stats_attrs(field)					\
> > > > > > +	do {								\
> > > > > > +		gb_loopback_ro_stats_attr(field, min, u);		\
> > > > > > +		gb_loopback_ro_stats_attr(field, max, u);		\
> > > > > > +		gb_loopback_ro_avg_attr(field);				\
> > > > > > +	} while (0)
> > > > > >
> > > > > >  #define gb_loopback_attr(field, type)					\
> > > > > >  static ssize_t field##_show(struct device *dev,				\
> > > > > > --
> > > > > > 2.31.1
> > > > > >
> > > > > >
> > > > >
> > > > > Did you test build this change?
> > > >
> > > > I built the module using make -C . M=drivers/staging/greybus to test
> > > > build it. I didn't get any errors.
> > >
> > > Really?  Can you provide the full build output for this file with your
> > > change?  I don't think you really built this file for the obvious
> > > reasons...
> >
> > I ran make -C . M=drivers/staging/greybus
> >
> > I got a three line output saying:
> > make: Entering directory '/work/linux'
> >   MODPOST drivers/staging/greybus//Module.symvers
> > make: Leaving directory '/work/linux'
> >
> > I just tried rebuilding the kernel with CONFIG_GREYBUS=m, and now I can
> > see what you are talking about. Why weren't these errors reported when I
> > ran the previous make command? Does that too check for the config
> > variables even when I specifically asked it to build a module?
>
> You were just asking it to build a subdirectory, not a specific
> individual file, and when you do that it looks at the configuration
> settings.
>

I see.

> It's always good to ensure that you actually build the files you modify
> before sending patches out.

Sorry, I googled about building a single module, and thought running
that command would have built it. Moreover, since the change was so
simple I didn't suspect anything when it got built correctly the first
time around.

I didn't look at how/where was the macro called and missed a very
obvious error. Now that I have looked at it, the only way I can think of
fixing this is changing the macro to a (inline?) function. Will
that be a desirable change?

And yes, I will definitely be more careful in the future.

>
> thanks,
>
> greg k-h