Re: [PATCH 04/11] exportfs: define _GNU_SOURCE for stat64

[Steve Dickson <SteveD@redhat.com>]

On 07/31/2014 07:34 AM, Jeff Layton wrote:
> On Thu, 31 Jul 2014 11:30:22 +1000
> NeilBrown <neilb@suse.de> wrote:
> 
>> On Wed, 30 Jul 2014 13:23:12 +0200 Natanael Copa <natanael.copa@gmail.com>
>> wrote:
>>
>>> Signed-off-by: Natanael Copa <ncopa@alpinelinux.org>
>>> ---
>>>  utils/exportfs/exportfs.c | 4 ++++
>>>  1 file changed, 4 insertions(+)
>>>
>>> diff --git a/utils/exportfs/exportfs.c b/utils/exportfs/exportfs.c
>>> index bf07555..7ab93d1 100644
>>> --- a/utils/exportfs/exportfs.c
>>> +++ b/utils/exportfs/exportfs.c
>>> @@ -12,6 +12,10 @@
>>>  #include <config.h>
>>>  #endif
>>>  
>>> +#ifndef _GNU_SOURCE
>>> +#define _GNU_SOURCE
>>> +#endif
>>> +
>>>  #include <sys/types.h>
>>>  #include <sys/stat.h>
>>>  #include <sys/vfs.h>
>>
>> These all look really sensible!
>>
>> One small suggestion: it would be really nice to see a comment in the code
>> explaining why _GNU_SOURCE is needed.  I suspect such comments get out of
>> date quickly, so maybe it wouldn't end up be all that useful.  But having a
>> comment is still, in my opinion, more useful than not.
>>
>> Thanks,
>> NeilBrown
>>
> 
> Agreed, they all look pretty sensible. That said, I think the supposed
> best way to enable _GNU_SOURCE on autoconf projects is to use
> AC_USE_SYSTEM_EXTENSIONS:
> 
>     https://www.gnu.org/software/autoconf/manual/autoconf-2.64/html_node/Posix-Variants.html
> 
Its sounds like these requests are mutually exclusive... Meaning
if we use the AC_USE_SYSTEM_EXTENSIONS we will not get the
comments as to why _GNU_SOURCE is needed because the these 
#ifndefs and #define will go away... which I guess is OK...

Natanael, would you mind incorporate AC_USE_SYSTEM_EXTENSIONS as well
as fix the typo Jeff pointed out in patch 11? 

Once that happens... I'm good to go on these... 

steved.