IRIX t1lib patch for 2.0

On IRIX, the order of the compile command matters a lot. If object file x depends on a symbol in library y, "-ly x.o" won't work. Also, using LDFLAGS to add library options, -lx, below is bad form. -- albert chin ([EMAIL PROTECTED]) -- snip snip --- libs/t1lib/t1lib.ac.origM

Replace klibtool with libtool

4:0:0 only to move away from the 3.x.y version currently used. Maybe something else is more suitable. -- albert chin ([EMAIL PROTECTED]) -- snip snip --- texk/make/programs.mk.orig Mon Feb 3 20:03:10 2003 +++ texk/make/programs.mk Mon Feb 3 20:24:46 2003 @@ -15,5 +15,6 @@ # When we

acspecific.m4 patch

1. AIX wants the return type of select to be int. 2. IRIX doesn't like the lack of a prototype in the AC_TRY_LINK line below. -- albert chin ([EMAIL PROTECTED]) -- snip snip --- texk/etc/autoconf/acspecific.m4.origMon Feb 3 18:47:42 2003 +++ texk/etc/autoconf/acspecific.m4 Mo

Re: Optimization problem with gcc 3.2.1 on Solaris 8

on Solaris 8 with no problems: $ cc -V cc: Sun WorkShop 6 update 2 C 5.3 Patch 111679-08 2002/05/09 -- albert chin ([EMAIL PROTECTED])

Re: `@' sign in path not treated properly

7;s enough to > recreate configure with never auto* tools. We also build with @sys (for a customer). Try the patch below. -- albert chin ([EMAIL PROTECTED]) -- snip snip --- texk/kpathsea/Makefile.in.orig Mon Feb 3 08:24:08 2003 +++ texk/kpathsea/Makefile.in Tue Feb 4 00:32:26 20

Re: licenses in teTeX

nline results to-date? That way, duplicate work is avoided. Even a mailing list would be good. -- albert chin ([EMAIL PROTECTED])