Mercurial > emacs
view src/s/isc3-0.h @ 3782:5b61a650f347
Changes for correct pgrp behavior from Bob Glickstein
<bobg@zindigo.z-code.com>:
* m/iris4d.h (LIB_STANDARD): Do include -lbsd here.
* s/irix4-0.h (USE_IRIX_BSDPGRP): #define this.
(GETPGRP_NO_ARG): Don't #define this.
* callproc.c [USE_IRIX_BSDPGRP] (Fcall_process): Call BSDsetpgrp
instead of plain setpgrp.
[USE_IRIX_BSDPGRP] (child_setup): Same.
* emacs.c [GETPGRP_NO_ARG] (main, shut_down_emacs): CPP clauses
deleted; this flag isn't needed anymore.
[USE_IRIX_BSDPGRP] (main, shut_down_emacs): Call BSDsetpgrp and
BSDgetpgrp instead of plain setpgrp and getpgrp.
* process.c [USE_IRIX_BSDPGRP] (create_process): Don't call
setpgrp before opening the tty. Call BSDsetpgrp instead of
setpgrp, after opening the tty.
* sysdep.c [USE_IRIX_BSDPGRP] (sys_suspend,
narrow_foreground_group, widen_foreground_group): Call BSDgetpgrp
and BSDsetpgrp instead of getpgrp and setpgrp.
author | Jim Blandy <jimb@redhat.com> |
---|---|
date | Wed, 16 Jun 1993 23:49:54 +0000 |
parents | 5de0e54e03ab |
children | 01de01b59c7f |
line wrap: on
line source
/* s- file for Interactive (ISC) Unix version 3.0 on the 386. */ #include "s/isc2-2.h" /* These have been moved into s-isc2-2.h. */ /* #define HAVE_SOCKETS #define HAVE_SELECT */ /* This appears on 3.0, presumably as part of what SunSoft call X2. */ #undef NO_X_DESTROY_DATABASE /* mt00@etherm.co.uk says this is needed for process.c. */ #define HAVE_TIMEVAL /* People say that using -traditional causes lossage with `const', so we might as well try getting rid of -traditional. */ #undef C_SWITCH_SYSTEM /* We indirectly #include s/usg5-3.h, which says to use libX11_s and libc_s. Martin Tomes <mt00@controls.eurotherm.co.uk> says that ISC has no libX11_s, and that linking with libc_s causes sbrk not to work. */ #undef LIB_X11_LIB #undef LIBX11_SYSTEM #define LIBX11_SYSTEM -lpt -lnls -lnsl_s -lc /* marko@tekelec.com (Marko Rauhamaa) says that his linker couldn't find memmove, but that sounds crazy - I thought all SYSV descendants had that. Let us know if this turns out to be wrong. */ #define memmove(d, s, n) safe_bcopy ((s), (d), (n))