Bug#711529: gnome-terminal: FTBFS on GNU/kFreeBSD

[Emilio Pozuelo Monfort]

On 11/06/13 11:55, Emilio Pozuelo Monfort wrote:
> On 11/06/13 11:43, Emilio Pozuelo Monfort wrote:
>> On 07/06/13 16:13, Petr Salinger wrote:
>>> Package: gnome-terminal
>>> Version: 3.8.2-1
>>> Severity: serious
>>> Tags: patch
>>> User: debian-bsd at lists.debian.org
>>> Usertags: kfreebsd
>>>
>>>
>>> Hi,
>>>
>>> the current version fails to build on GNU/kFreeBSD.
>>> The F_DUPFD_CLOEXEC is not widespread fcntl, see also changes between
>>> http://pubs.opengroup.org/onlinepubs/009695399/functions/fcntl.html
>>> http://pubs.opengroup.org/onlinepubs/9699919799/functions/fcntl.html
>>
>> Seems like a POSIX.2008 addition.
>>
>>> Please allow also fallback variant.
>>
>> The patch seems fine but given that F_DUPFD_CLOEXEC has been implemented in
>> FreeBSD[1], I wonder if the __linux__ test shouldn't be changed to a
>> HAVE_F_DUPFD_CLOEXEC check (or to a hurd one). How long do you think it'll take
>> for Debian's freebsd kernel to have that?
> 
> Actually Hurd has had support for F_DUPFD_CLOEXEC for a while, so this should
> either be __FreeBSD_kernel__ (or whatever) or a HAVE_F_DUPFD_CLOEXEC (with the
> appropriate configure check) or perhaps even an #ifdef F_DUPFD_CLOEXEC (assuming
> it will be a define in kfreebsd, it is one on linux and hurd).

Given the following from [1]

"The <fcntl.h> header shall define the following symbolic constants for the cmd
argument used by fcntl(). The values shall be unique and shall be suitable for
use in #if preprocessing directives"

I say we can just do an #ifdef F_DUPFD_CLOEXEC

I'll adjust the patch for that and upload.

Cheers,
Emilio

[1] http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/fcntl.h.html