steve@lia (Stephen Williams) (06/30/91)
I've been having some trouble with Andrew patch level 10 on SunOS 4.1.1. The pertinent stuff: SunOS 4.1.1 SPARCstation IPC X11R4 server (Xsun) on a color (cgtree) monitor There are several problems. First, with compiling: When compiling tabs.do, the doindex command fails with the message doindex: indexing tabs.do ...ld.so: map heap error (9) for /dev/zero It is interesting that I can run doindex on all the individual .do files and I get no errors. I have the same trouble with the console tool: /local/andrew/bin/doindex -d /local/andrew/dlib/atk consolea.do doindex: indexing consolea.do ...ld.so: map heap error (9) for /dev/zero So I go without the console and I try some of the tools. The first problem I notice is that when I try to use a menu, the cursor (the -> cursor) is wrong. It is displayed as a white box with a transparent arrow. It seems that the X cursor mask bits are reversed. Last problem (for today): typescript doesn't work. It didn't work before patch 10, but patch 10 at work (4.1 on a SPARCstation 1) does work. The symptoms are the same as before patch 10: The child process (usually /bin/csh) dies immediately (sometimes after printing the prompt. I am having none of these troubles on a SPARCstation with the X11R4 server (Xsun), a monochrome monitor, and SunOS 4.1. I am frustrated. It is teasing me! --Steve Williams steve@lia.com
dl2n+@ANDREW.CMU.EDU (Daniel Edward Lovinger) (07/01/91)
fernwood!lia!steve@uunet.uu.net (Stephen Williams) writes: > When compiling tabs.do, the doindex command fails with > the message > doindex: indexing tabs.do ...ld.so: map heap error (9) for /dev/zero > It is interesting that I can run doindex on all the > individual .do files and I get no errors. I had exactly the same problem and spent about 12 hours banging my head against a wall until I realized that the top level Makefile hadn't been generated on a Sparc. Look at the top level Makefile, and if the first three macros aren't SYSTEM_H_FILE = sun4_41/system.h SYS_IDENT = sun4_41 SYS_OS_ARCH = sun_sparc_41 then you will need to recreate it - the bogus values were be enshrined since the top level Makefile installs system.h. imake -Timake.tmpl -I. -I./config -DTOPDIR=. (and of course, check to make sure it thinks you have a sun4_41 again) You should have no problems after doing this. I didn't ... Dan Lovinger Computing & Comunications Carnegie Mellon U. Internet: dl2n+@andrew.cmu.edu Bitnet: dl2n+%andrew.cmu.edu@carnegie "... and in the stillness, they heard the cry of the golden banana ..." ------------------------------------------------------------------------------