[lsb-discuss] sysconf(_SC_2_C_VERSION) return -1

Andrew Josey ajosey at rdg.opengroup.org
Tue Aug 22 23:49:43 PDT 2006


I believe that the first symbol (_SC_2_C_VERSION) is no longer required
in latest standards (it was removed in a recent technical corrigenda
IEEE Std 1003.1-2001/Cor 2-2004, item XBD/TC2/D6/32 ). So its ok not to
support the symbol.

The other flags are all supported and ok, i think the test reports the
status  of all these flags all in one go, so no problem there.

For FIPS sign off, just saying that the results are correct would be
fine. We could update this test to no longer flag the issue -- under the
LSB_TEST setting (as other modes have to remain intact as per the
license). It is a FIP since for the underlying POSIX-1990/1996 tests it
was optional whether the system supported the POSIX.2 C Language
Binding.

I put a page up in the wiki based on FIPS in the runtime tests that i
hope will be helpful.
http://www.freestandards.org/en/FIPsResults
regards
Andrew

On Tue, 2006-08-22 at 15:02 -0400, Marc Tardif wrote:
> I'm using Ubuntu 6.06 LTS and running the LSB Runtime Tests 3.1.0-3
> for
> i486. The problem is that I'm encountering the following FIP results:
> 
>     sysconf(_SC_2_C_VERSION) returned -1
>        - indicating the feature group is not (fully) supported
>     sysconf(_SC_XOPEN_ENH_I18N) returned 1
>        - indicating the feature group is fully supported
>     sysconf(_SC_XOPEN_CRYPT) returned 1
>        - indicating the feature group is supported
>     sysconf(_SC_XOPEN_SHM) returned 1
>        - indicating the feature group is supported
>     sysconf(_SC_XOPEN_LEGACY) returned 1
>        - indicating the feature group is fully supported
> 
> First, what could be the reason(s) for _SC_2_C_VERSION not being fully
> supported?
> 
> Second, what is wrong with the other return values being 1?
> 




More information about the lsb-discuss mailing list