The problem with adding _BSD_SOURCE globally is that it's too easy for people to not be aware of their use of non-compliant routines unless they get a per-module warning which they can then document by including such overrides themselves. So I agree that we want c99 to be set globally, but I'd prefer to see the overrides done on per-module basis.
On Fri, Dec 12, 2014 at 7:02 AM, Mike Holmes <[email protected]> wrote: > > > > On 12 December 2014 at 07:48, Maxim Uvarov <[email protected]> > wrote: >> >> On 12/12/2014 03:38 PM, Bill Fischofer wrote: >> >>> Are we really ready to do this? I thought the idea was we were going to >>> have each routine that needed it include the BSD/POSIX #defines so that >>> they document their need for this on a case-by-case basis. >>> >> >> Might be it was better to send this as RFC as discussion patch. >> > > I think we need to wait until the code catches up and this CI job > <https://ci.linaro.org/view/odp-ci/job/odp-api-check-native-c99/buildhw=x86_64,label=build/19/console> > runs clean, then turning on c99 will enforce it going forward. > > If we want to change what our future flags will be, lets add them to this > job first. > > Obviously -std=c99 is a no brainer addition and it there already, > however the blanket adding of _BSD_SOURCE may not be the best idea. > I think I saw a way on a forum somewhere that we can add this flags effect > per API used - that way we would better document what we depend on for > linux-generic - if there is no such mechanism lets add _BSD_SOURCE to the > CI job. > > >> Most of linux-generic implementation will need posix defines. I'm not >> sure if that 2 options overlap each other or not. But I think all code >> should follow the same posix standard. >> > >> Maxim. >> >> >>> On Fri, Dec 12, 2014 at 6:31 AM, Maxim Uvarov <[email protected] >>> <mailto:[email protected]>> wrote: >>> >>> Define _POSIX_C_SOURCE and _BSD_SOURCE to make gcc accept >>> system calls and turn on -std=c99. >>> >>> Signed-off-by: Maxim Uvarov <[email protected] >>> <mailto:[email protected]>> >>> --- >>> configure.ac <http://configure.ac> | 2 ++ >>> 1 file changed, 2 insertions(+) >>> >>> diff --git a/configure.ac <http://configure.ac> b/configure.ac >>> <http://configure.ac> >>> index 2f27261..d9167a5 100644 >>> --- a/configure.ac <http://configure.ac> >>> +++ b/configure.ac <http://configure.ac> >>> @@ -179,6 +179,8 @@ ODP_CFLAGS="$ODP_CFLAGS -Wmissing-declarations >>> -Wold-style-definition -Wpointer- >>> ODP_CFLAGS="$ODP_CFLAGS -Wcast-align -Wnested-externs -Wcast-qual >>> -Wformat-nonliteral" >>> ODP_CFLAGS="$ODP_CFLAGS -Wformat-security -Wundef -Wwrite-strings" >>> >>> +ODP_CFLAGS="$ODP_CFLAGS -std=c99 -D_BSD_SOURCE >>> -D_POSIX_C_SOURCE=200809L" >>> + >>> ############################################################ >>> ############## >>> # Default include setup >>> ############################################################ >>> ############## >>> -- >>> 1.8.5.1.163.gd7aced9 >>> >>> >>> _______________________________________________ >>> lng-odp mailing list >>> [email protected] <mailto:[email protected]> >>> http://lists.linaro.org/mailman/listinfo/lng-odp >>> >>> >> >> _______________________________________________ >> lng-odp mailing list >> [email protected] >> http://lists.linaro.org/mailman/listinfo/lng-odp >> > > > -- > *Mike Holmes* > Linaro Sr Technical Manager > LNG - ODP >
_______________________________________________ lng-odp mailing list [email protected] http://lists.linaro.org/mailman/listinfo/lng-odp
