From JAR@FED.ORNL.GOV Tue May 6 16:19:23 1997 Received: from MIT.EDU (PACIFIC-CARRIER-ANNEX.MIT.EDU [18.69.0.28]) by rt-11.MIT.EDU (8.7.5/8.7.3) with SMTP id QAA07790 for ; Tue, 6 May 1997 16:19:19 -0400 Received: from fed.ornl.gov by MIT.EDU with SMTP id AA05138; Tue, 6 May 97 15:18:42 EST Received: from FED.ORNL.GOV by FED.ORNL.GOV (PMDF V5.0-6 #12522) id <01IIK0F40FRQ8X1GN7@FED.ORNL.GOV> for krb5-bugs@mit.edu; Tue, 06 May 1997 16:19:14 -0400 (EDT) Message-Id: <01IIK0F40GQ08X1GN7@FED.ORNL.GOV> Date: Tue, 06 May 1997 16:19:14 -0400 (EDT) From: jar@ornl.gov Sender: "Jim Rome (423) 574-1306 E-mail: jar@ornl.gov" To: krb5-bugs@MIT.EDU Subject: FWD: Bad error in the K5 config file >Number: 424 >Category: krb5-build >Synopsis: FWD: Bad error in the K5 config file >Confidential: no >Severity: serious >Priority: medium >Responsible: tlyu >State: closed >Class: sw-bug >Submitter-Id: unknown >Arrival-Date: Tue May 06 16:20:00 EDT 1997 >Last-Modified: Mon Aug 11 23:00:26 EDT 1997 >Originator: >Organization: >Release: >Environment: >Description: >How-To-Repeat: >Fix: >Audit-Trail: From: "Theodore Y. Ts'o" To: krb5-bugs@MIT.EDU, jar@ornl.gov Cc: gnats-admin@RT-11.MIT.EDU, krb5-prs@RT-11.MIT.EDU Subject: Re: pending/424: FWD: Bad error in the K5 config file Date: Wed, 7 May 1997 09:53:40 -0400 Date: Tue, 06 May 1997 16:19:14 -0400 (EDT) From: jar@ornl.gov There is a problem in the config files for krb5p1. In src/util/profile/configure, around line 1199, the code has a comment in it that gets catted to a C program which doesn't use # for comments or recognizes #line as a valid preprocessor command. I'm not sure I understand your bug report. What platform and C compiler were you using for which this was a problem? For the vast majority of the platforms that I'm familiar with, #line isn't a problem. Also, the configure script uses "#line" in literally hundreds of places all over the Kerberos V5 tree, not just in the line which you reported. Was this change sufficient to make the tree work for you? And if so, do you have any insight why #line wasn't causing you problems elsewhere? (Search for #line in any configure script and you'll see it all over the place.) - Ted Responsible-Changed-From-To: gnats-admin->tlyu Responsible-Changed-By: tlyu Responsible-Changed-When: Wed Jun 18 17:59:53 1997 Responsible-Changed-Why: Refiled State-Changed-From-To: open-feedback State-Changed-By: tlyu State-Changed-When: Wed Jun 18 18:00:44 1997 State-Changed-Why: user error? State-Changed-From-To: feedback-closed State-Changed-By: tlyu State-Changed-When: Mon Aug 11 23:00:09 1997 State-Changed-Why: No response; punt. >Unformatted: There is a problem in the config files for krb5p1. In src/util/profile/configure, around line 1199, the code has a comment in it that gets catted to a C program which doesn't use # for comments or recognizes #line as a valid preprocessor command. Things like this make it hard to deploy Kerberos. I haven't gotten make to work yet on an HP. I had several bad include files on my AIX 4.2 machine. See my Web page: http://www.epm.ornl.gov/~jar/HowToKerb.htm > > # If we cannot run a trivial program, we must be cross compiling. > echo $ac_n "checking whether cross-compiling""... $ac_c" 1>&6 > if eval "test \"`echo '$''{'ac_cv_c_cross'+set}'`\" = set"; then > echo $ac_n "(cached) $ac_c" 1>&6 > else > if test "$cross_compiling" = yes; then > ac_cv_c_cross=yes > else > cat > conftest.$ac_ext < /*#line 1200 "configure" ########## BUG######## */ > #include "confdefs.h" > main(){return(0);} > EOF > { (eval echo configure:1204: \"$ac_link\") 1>&5; (eval $ac_link) 2>&5; } > if test -s conftest && (./conftest; exit) 2>/dev/null; then > ac_cv_c_cross=no > else > ac_cv_c_cross=yes > fi > fi > rm -fr conftest* > fi > James Rome ORNL