skrenta@eecs.nwu.edu (Richard Skrenta) (10/25/88)
Just a simple question here. I'm trying to get Micro Gnu Emacs to work under 286 Xenix 2.0.4. It compiles fine and makes a 67K executable. When I try to run it, however, it says "mg: too big". I've run programs much bigger than this, and I can't find any reference to that error message in my manuals. Does anyone know what "too big" refers to here? Thanks Rich Skrenta
debra@alice.UUCP (Paul De Bra) (10/26/88)
In article <3700006@eecs.nwu.edu> skrenta@eecs.nwu.edu (Richard Skrenta) writes: >Just a simple question here. I'm trying to get Micro Gnu Emacs to work >under 286 Xenix 2.0.4. It compiles fine and makes a 67K executable. When >I try to run it, however, it says "mg: too big"... "too big" could mean that the program is too big (you should try "size" too find out how big it is on startup, not ls -l), but it can also mean that you tried to put more than 64k in one segment, without the loader noticing. (this is possible because the loader does not take the stack into account) Paul. -- ------------------------------------------------------------------------- |debra@research.att.com | uunet!research!debra | att!grumpy!debra | -------------------------------------------------------------------------
root@conexch.UUCP (Larry Dighera) (10/26/88)
In article <3700006@eecs.nwu.edu> skrenta@eecs.nwu.edu (Richard Skrenta) writes: >Just a simple question here. I'm trying to get Micro Gnu Emacs to work >under 286 Xenix 2.0.4. It compiles fine and makes a 67K executable. When >I try to run it, however, it says "mg: too big". I've run programs much >bigger than this, and I can't find any reference to that error message in >my manuals. Does anyone know what "too big" refers to here? I don't have the answer to this one, but I can shed a little light on the subject. Since installing VP/ix on conexch I have been having similar errors when attempting to run, of all things, uustat, and attempting to run spell results in: table = malloc(0xc848) fails sbrk(0) = 0x3868 spell: cannot initialize hash table This is with Xenix 386 2.3.0. I noticed that vpixadm added two additional PATH entries to an already long PATH string. So, first I removed several shell functions from my environment; no change. Next, I shortened the PATH to just /bin:/usr/bin. Success! All runs as before. I have no explanation for this. It is annoying. Larry Dighera -- USPS: The Consultants' Exchange, PO Box 12100, Santa Ana, CA 92712 TELE: (714) 842-6348: BBS (N81); (714) 842-5851: Xenix guest account (E71) UUCP: conexch Any ACU 2400 17148425851 ogin:-""-ogin:-""-ogin: nuucp UUCP: ...!uunet!turnkey!conexch!root || ...!trwrb!ucla-an!conexch!root
chip@vector.UUCP (Chip Rosenthal) (10/27/88)
In article <3700006@eecs.nwu.edu> skrenta@eecs.nwu.edu (Richard Skrenta) writes: >I try to run it, however, it says "mg: too big". >Does anyone know what "too big" refers to here? Yeah...your initialized data/stack segment is greater than 64K. Well, you might ask, why does the compiler allow such a thing to be created? The reason is that XENIX 286 uses a fixed stack which is allocated at run time. It is only at this time that it can see that the thing won't shoehorn into memory. The answer is obvious: decrease your stack size or decrease the amount of static data in this segment. The former may be done with the fixhdr(C) command, which allows you to specify a stack size. The latter may be done by compiling with the -Mt flag. For example, -Mt32 says any data items larger than 32 bytes should be moved into another segment. Of course, you must be using either the large or the huge memory model to do this. -- Chip Rosenthal chip@vector.UUCP | I've been a wizard since my childhood. Dallas Semiconductor 214-450-0486 | And I've earned some respect for my art.