Re: [PATCH] staging: unisys: visorbus: constify attribute_group structures.

[Greg KH <greg@kroah.com>]

On Mon, Jul 17, 2017 at 09:36:58PM +0000, Kershner, David A wrote:
> > -----Original Message-----
> > From: Greg KH [mailto:greg@kroah.com]
> > Sent: Monday, July 17, 2017 8:38 AM
> > To: Arvind Yadav <arvind.yadav.cs@gmail.com>
> > Cc: Kershner, David A <David.Kershner@unisys.com>; Sell, Timothy C
> > <Timothy.Sell@unisys.com>; Thompson, Bryan E.
> > <bryan.thompson@unisys.com>; jon.frisch@unisys.com; Binder, David
> > Anthony <David.Binder@unisys.com>; *S-Par-Maintainer
> > <SParMaintainer@unisys.com>; devel@driverdev.osuosl.org; linux-
> > kernel@vger.kernel.org
> > Subject: Re: [PATCH] staging: unisys: visorbus: constify attribute_group
> > structures.

Why is this in your email body?

> > 
> > On Mon, Jul 17, 2017 at 05:43:14PM +0530, Arvind Yadav wrote:
> > > Hi Greg,
> > >
> > >
> > > On Monday 17 July 2017 04:15 PM, Greg KH wrote:
> > > > On Mon, Jul 17, 2017 at 02:55:37PM +0530, Arvind Yadav wrote:
> > > > > attribute_groups are not supposed to change at runtime. All functions
> > > > > working with attribute_groups provided by <linux/sysfs.h> work
> > > > > with const attribute_group. So mark the non-const structs as const.
> > > > >
> > > > > Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
> > > > > ---
> > > > >   drivers/staging/unisys/visorbus/visorbus_main.c | 4 ++--
> > > > >   drivers/staging/unisys/visorbus/visorchipset.c  | 2 +-
> > > > >   2 files changed, 3 insertions(+), 3 deletions(-)
> > > > Why not just use the ATTRIBUTE_GROUPS() macro for these?  Or is there
> > > > something that is preventing that?
> > > Yes, we can use. if we are only initializing '.attrs'.
> > > ATTRIBUTE_GROUPS() will not work if we will initialize other member of
> > > attribute_group like 'bin_attrs', 'is_visible', and  'name'.
> > 
> > That means you should redo this patch :)
> > 
> > Also, your changelog text had a typo, it is "attribute_group", not
> > "attribute_groups".
> > 
> 
> Greg, are you recommending that we shouldn't be setting the attribute_group
> .name field? What does it pick up if we don't specify it?

Why do you want a name for your group?  Anyway, yes, you are right, if
you set a .name, then you can't use the macro, my fault, I hadn't looked
at it in a long time.

> Also, for our attribute_groups in visorchipset, we are defining it with two
> different attribute_group variables. Are you allowed to use two different
> attribute_group variables in an attribute_groups, or is this frowned upon and
> we should flatten it out to just one? An example that we used in the kernel was:
> 
> 	static const struct attribute_group *l2_cache_pmu_attr_grps[] = {
> 		&l2_cache_pmu_format_group,
> 		&l2_cache_pmu_cpumask_group,
> 		NULL,
> 	};

Nah, that's fine, sorry for the noise.  But the changelog text still
should be fixed so I can take this.

thanks,

greg k-h