Re: [GIT PULL] hash addresses printed with %p

[Dave Young]

On 12/03/17 at 06:33pm, Joe Perches wrote:
> On Mon, 2017-12-04 at 10:02 +0800, Dave Young wrote:
> > I think 0400 is good enough for this issue.
> > 
> > Greg, would you like to agree add an extra macro like below?
> []
> > -static struct map_attribute map_type_attr = __ATTR_RO(type);
> > -static struct map_attribute map_phys_addr_attr   = __ATTR_RO(phys_addr);
> > -static struct map_attribute map_virt_addr_attr  = __ATTR_RO(virt_addr);
> > -static struct map_attribute map_num_pages_attr  = __ATTR_RO(num_pages);
> > -static struct map_attribute map_attribute_attr  = __ATTR_RO(attribute);
> > +static struct map_attribute map_type_attr = __ATTR_IRUSR(type);
> > +static struct map_attribute map_phys_addr_attr   = __ATTR_IRUSR(phys_addr);
> > +static struct map_attribute map_virt_addr_attr  = __ATTR_IRUSR(virt_addr);
> > +static struct map_attribute map_num_pages_attr  = __ATTR_IRUSR(num_pages);
> > +static struct map_attribute map_attribute_attr  = __ATTR_IRUSR(attribute);
> >  
> >  /*
> >   * These are default attributes that are added for every memmap entry.
> > --- linux-x86.orig/include/linux/sysfs.h
> > +++ linux-x86/include/linux/sysfs.h
> > @@ -112,6 +112,11 @@ struct attribute_group {
> >  	.store	= _store,						\
> >  }
> >  
> > +#define __ATTR_IRUSR(_name) {					
> 
> I'd much prefer __ATTR_0400(_name)
> 

I'm also fine with above, easier to get the meaning.

Thanks for the suggestion.
Dave