Mail Archives: cygwin/2004/03/24/11:17:48
On Wed, Mar 24, 2004 at 08:18:11PM +0000, Ghanshyam wrote:
>I found some problem in sysconf() system call with respect to following
>assertions defined in "IEEE std 2003.1 -1992 Test methods for measuring
>conformance to POSIX-part1" document
>
>
>10(A) If STREAM_MAX is defined when the header <limits.h> is included:
> A call to sysconf(_SC_STREAM_MAX) either returns -1 without changing errno
> or returns a value greater than or equal to {STREAM_MAX}.
>Otherwise:
> A call to sysconf(_SC_STREAM_MAX) either returns -1 without changing errno
> or returns a value greater than or equal to {_POSIX_STREAM_MAX}.
>
>11(A) If TZNAME_MAX is defined when the header <limits.h> is included:
> A call to sysconf(_SC_TZNAME_MAX) either returns -1 without changing errno
> or returns a value greater than or equal to {TZNAME_MAX}.
>Otherwise:
> A call to sysconf(_SC_TZNAME_MAX) either returns -1 without changing errno
> or returns a value greater than or equal to {_POSIX_TZNAME_MAX}.
>
>********when called with argument _SC_STREAM_MAX or _SC_TZNAME_MAX, It
>returns -1 but set errno to 22 (EINVAL) but is should not modify its
>value.
>
>By looking into code, I found that sysconf() is not implemented for these
>argument values!
Please calm down.
>Is any one going to address these in near future?
Sorry, no. Were you under the impression that cygwin is guaranteed to
conform to some posix standard or that people would be actively fixing
any problems reported in posix compliance? Unfortunately, neither is
true.
As I've previously noted, you could always contribute patches yourself.
Alternately, (and I know that Mentor knows about this) you could engage
the services of Red Hat or some other organization to implement whatever
you need.
cgf
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
- Raw text -