defining HAVE_SIGNAL_H if the header is available.

Added a check in configure that tests if the sig_atomic_t type is
available, defining HAVE_SIG_ATOMIC_T if it is available. Providing
a suitable default in setup_once.h if not available.

Added a check in configure that tests if the sig_atomic_t type is
already defined as volatile, defining HAVE_SIG_ATOMIC_T_VOLATILE
if it is available and already defined as volatile.

Reintroduce checking for HAVE_MSG_NOSIGNAL in configure script, so that we don't depend on header inclusion order for a valid check.

Define NEED_MALLOC_H if including <stdlib.h> is not enough for proper compilation and <malloc.h> must also be included.

libtool cross-compiled on linux with mingw32

things such as C++ compiler actually is a bad thing and since we don't need
that detection I added a work-around, much inspired by a previous patch by
Paolo Bonzini. This also shortens the configure script quite a lot.

moved the curl_off_t check to within the --enable-debug block where it belongs since it is a somewhat ugly hack

done in one single function CURL_CHECK_FUNC_GETNAMEINFO which
will only define HAVE_GETNAMEINFO if the function has been found
AND the type of its arguments has been properly been detected

CURL_FUNC_GETNAMEINFO_ARGTYPES now also checks first argument.
All related changes taken to cares configuration scripts.

ifdefs where . was used instead of _!

in mingw builds

macros accordingly.

way of detecting it

Added constants that will be used by ares_getaddrinfo. Made ares_getnameinfo use the reentrant getservbyport (getservbyport_r) if it isavailable to ensure it works properly in a threaded environment

inventing and providing our own. Hopefully this solves a HP-UX 11.00 problem.

  Make UDP sockets non-blocking. I've confirmed that at least on Linux 2.4 a
  read event can come back from poll() on a valid SOCK_DGRAM socket but
  recv(2) will still block. This patch doesn't ignore EAGAIN in
  read_udp_packets(), though maybe it should. (This patch was edited by Daniel
  Stenberg and a new configure test was added (imported from curl's configure)
  to properly detect what non-blocking socket approach to use.)

on Solaris and FreeBSD. At least.

include, as it seems at least some AIX versions don't really allow it to be
include at the same time as the original nameser.h.

(IRIX 6.5.22 it seems)

Provided implementations of inet_net_pton and inet_pton from BIND for systems that do not include these functions. These will be necessary for CIDR support and IPv6 support.