- Merge 4.4BSD strerror.3 man page changes that summarize information
for all the error printing functions. This makes the perror(3) page
obsolete.
- Implement all error functions in terms of strerror_r(), including
__strerror() which is not used internally. Can it be removed?
- Bump version for strerror_r
This is a workaround to make gcc's behavior consistent, since gcc can
transform:
printf("%s\n", s) -> puts(s)
fprintf(fp, "%s", s) -> fputs(s, fp)
as an optimization.
I've left the _DIAGASSERT's that make sure that s != NULL alone because we
should really still abort in a debugging environment.
* Make vfprintf_unlocked() an internal function, c.f. __svfscanf_unlocked().
* Add internal names for arc4random(), endnetpath(), fhstatvfs(),
fstatvfs(), mkstemp(), shquote(), statvfs(), taddr2uaddr(), uaddr2taddr(),
uuid_create_nil(), uuid_is_nil(), and wcwidth().
* Include namespace.h where supposed to.
_NETBSD_SOURCE as this makes cross building from older/newer versions of
NetBSD harder, not easier (and also makes the resulting tools 'different')
Wrap all required code with the inclusion of nbtool_config.h, attempt to
only use POSIX code in all places (or when reasonable test w. configure and
provide definitions: ala u_int, etc).
Reviewed by lukem. Tested on FreeBSD 4.9, Redhat Linux ES3, NetBSD 1.6.2 x86
NetBSD current (x86 and amd64) and Solaris 9.
Fixes PR's: PR#17762 PR#25944
Its API in inherently non-threadsafe (and it depends on certain properties
of the underlying stdio implementation), so it shouldn't be a first-class
function for general use, polluting the application namespace.
Also remove the FLOCKFILE() code from it - this is under control of the
calling function now.
(XXX I'm not sure whether the FLOCKFILE() should be kept in fgetln(),
this function cannot be used in multiple threads anyway. It doesn't
hurt much, and it might prevent corruption of internal FILE structures.)
for SCNd8 and friends - so we should support it properly (and not only
by pure luck on alignment blind little endian archs - it used the
"h" code path, storing a 16bit int via a byte pointer).
This makes sparc64 pass regress/lib/libc/int_fmtio.
* Rename "config.h" to "nbtool_config.h" and
HAVE_CONFIG_H to HAVE_NBTOOL_CONFIG_H.
This makes in more obvious in the source when we're using
tools/compat/config.h versus "standard autoconf" config.h
* Consistently move the inclusion of nbtool_config.h to before
<sys/cdefs.h> so that the former can provide __RCSID() (et al),
and there's no need to protect those macros any more.
These changes should make it easier to "tool-ify" a program by adding:
#if HAVE_NBTOOL_CONFIG_H
#include "nbtool_config.h"
#endif
to the top of the source files (for the general case).
