--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/src/s/irix5-0.h	Mon Aug 13 08:45:50 2007 +0200
@@ -0,0 +1,124 @@
+/* Synched up with: FSF 19.31. */
+
+#include "usg5-4.h"
+
+#define IRIX5
+
+#ifdef LIBS_SYSTEM
+#undef LIBS_SYSTEM
+#endif
+
+#ifdef LIB_STANDARD
+#undef LIB_STANDARD
+#endif
+
+#ifdef SYSTEM_TYPE
+#undef SYSTEM_TYPE
+#endif
+#define SYSTEM_TYPE "irix"
+
+#ifdef SETUP_SLAVE_PTY
+#undef SETUP_SLAVE_PTY
+#endif
+
+/* thomas@mathematik.uni-bremen.de says this is needed.  */
+/* Make process_send_signal work by "typing" a signal character on the pty.  */
+#define SIGNALS_VIA_CHARACTERS
+
+/* Define HAVE_ALLOCA to say that the system provides a properly
+   working alloca function and it should be used. */
+#define HAVE_ALLOCA
+#undef C_ALLOCA
+
+#ifndef NOT_C_CODE
+#ifndef __GNUC__
+#include <alloca.h>
+#endif
+#endif
+
+/* SGI has all the fancy wait stuff, but we can't include sys/wait.h
+   because it defines BIG_ENDIAN and LITTLE_ENDIAN (ugh!.)  Instead
+   we'll just define WNOHANG right here.
+   (An implicit decl is good enough for wait3.)  */
+/* [XEmacs: Now that we don't use BIG_ENDIAN/LITTLE_ENDIAN, it's safe
+   to include wait.h.  Should something change here?] */
+
+/* #define WNOHANG		0x1 */
+
+/* No need to use sprintf to get the tty name--we get that from _getpty.  */
+#ifdef PTY_TTY_NAME_SPRINTF
+#undef PTY_TTY_NAME_SPRINTF
+#endif
+#define PTY_TTY_NAME_SPRINTF
+/* No need to get the pty name at all.  */
+#ifdef PTY_NAME_SPRINTF
+#undef PTY_NAME_SPRINTF
+#endif
+#define PTY_NAME_SPRINTF
+#ifdef emacs
+char *_getpty ();
+#endif
+/* We need only try once to open a pty.  */
+#define PTY_ITERATION
+/* Here is how to do it.  */
+/* It is necessary to prevent SIGCHLD signals within _getpty.
+   So we block them. */
+#define PTY_OPEN						\
+{								\
+  char *name;							\
+  EMACS_BLOCK_SIGCHLD;						\
+  name = _getpty (&fd, O_RDWR | O_NDELAY, 0600, 0);		\
+  EMACS_UNBLOCK_SIGCHLD;					\
+  if (name == 0)						\
+    return -1;							\
+  if (fd < 0)							\
+    return -1;							\
+  if (fstat (fd, &stb) < 0)					\
+    return -1;							\
+  strcpy (pty_name, name);					\
+}
+
+/* jpff@maths.bath.ac.uk reports `struct exception' is not defined
+   on this system, so inhibit use of matherr.  */
+#define NO_MATHERR
+
+/* Info from simon@lia.di.epfl.ch (Simon Leinen) suggests this is needed.  */
+#define GETPGRP_NO_ARG
+
+/* Ulimit(UL_GMEMLIM) is busted...  */
+#define ULIMIT_BREAK_VALUE 0x14000000
+
+/* Tell process_send_signal to use VSUSP instead of VSWTCH.  */
+#define PREFER_VSUSP
+
+/* Because unexelfsgi.c cannot handle a ".sbss" section yet, we must
+   tell the linker to avoid making one.  SGI's cc does this by
+   default, but GCC (at least 2.5.8 and 2.6.0) doesn't. */
+#ifdef __GNUC__
+#define LD_SWITCH_SYSTEM -G 0
+#endif
+
+/* define MAIL_USE_FLOCK if the mailer uses flock
+   to interlock access to /usr/spool/mail/$USER.
+   The alternative is that a lock file named
+   /usr/spool/mail/$USER.lock.  */
+
+#define MAIL_USE_FLOCK
+
+/* use K&R C */
+/* XEmacs change -- use ANSI, not K&R */
+#ifndef __GNUC__
+#define C_SWITCH_SYSTEM -xansi
+#endif
+
+/* jackr@engr.sgi.com says that you can't mix different kinds of
+   signal-handling functions under IRIX 5.3.  I'm going to assume
+   that that was the reason this got broken.  Now that the
+   signal routines are fixed up, maybe this will work. --ben */
+/* Nope, it doesn't.  I've tried lots of things; it must be
+   genuinely broken. */
+/* XEmacs addition: People on IRIX 5.2 and IRIX 5.3 systems have
+   reported that they can't break out of (while t) using C-g or C-G.
+   This does not occur on other systems, so let's assume that SIGIO
+   is broken on these systems. */
+#define BROKEN_SIGIO