On Wed, Oct 27, 1999 at 03:50:05PM +0200, Hrvoje Niksic wrote: Did you use --pdump, and if yes did you remove the src/xemacs.dmp file you may already have? I didn't want to muck with the makefiles yet (to add a "don't load the xemacs.dmp file no matter what" parameter). OG.

Hrvoje Niksic <hniksic(a)pc-hrvoje.srce.hr&gt; writes: You are using Debian potato aren't you? After I upgraded glibc package to glib-2.1.2-6 I can no longer compile XEmacs without --with-system-malloc (vanilla XEmacs 21.2.7, 21.2.19, CVS version of 21.2 and 21.1). Backtrace for my case follows: gdb ./src/temacs ./src/core #0 0x8152ff9 in Fboundp (sym=0) at symbols.c:485 485 CHECK_SYMBOL (sym); (gdb) where #0 0x8152ff9 in Fboundp (sym=0) at symbols.c:485 #1 0x8082742 in xmalloc (size=16) at alloc.c:376 #2 0x80a691c in sort_args (argc=4, argv=0xbffffa74) at emacs.c:1810 #3 0x80a5e00 in xemacs_21_1_7_i686_pc_linux (argc=4, argv=0xbffffa74, envp=0xbffffa88, restart=0) at emacs.c:572 #4 0x80a80d3 in main (argc=4, argv=0xbffffa74, envp=0xbffffa88) at emacs.c:2069 gdb ./temacs (gdb) break xmalloc Breakpoint 1 at 0x80825a3: file alloc.c, line 392. (gdb) run Starting program: /home/penny/xemacs-21.1.7/src/./temacs Breakpoint 1, xmalloc (size=4) at alloc.c:392 392 void *val = (void *) malloc (size); (gdb) n 394 if (!val && (size != 0)) memory_full (); (gdb) p val $1 = (void *) 0x0 I don't know why, but malloc returns null pointer. FWIW, I can compile and run FSF Emacs with Doug Lea malloc without a problem. It's either Debian people did the wrong fix or XEmacs' usage of Doug Lea malloc is wrong. -- Yoshiki Hayashi