Skip Menu |
 

Download (untitled) / with headers
text/plain 1.7KiB
From evanc@synapse.net Sun Nov 30 09:31:39 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 JAA07580 for <bugs@RT-11.MIT.EDU>; Sun, 30 Nov 1997 09:31:38 -0500
Received: from piano.synapse.net by MIT.EDU with SMTP
id AA23681; Sun, 30 Nov 97 09:31:37 EST
Received: (qmail 1886 invoked by uid 1001); 30 Nov 1997 14:31:32 -0000
Message-Id: <19971130143132.1885.qmail@piano.synapse.net>
Date: 30 Nov 1997 14:31:32 -0000
From: evanc@synapse.net
To: krb5-bugs@MIT.EDU
Subject: FreeBSD already has libcom_err.a
X-Send-Pr-Version: 3.99

Show quoted text
>Number: 508
>Category: krb5-libs
>Synopsis: FreeBSD already has a libcom_err.a
>Confidential: no
>Severity: non-critical
>Priority: medium
>Responsible: krb5-unassigned
>State: open
>Class: change-request
>Submitter-Id: unknown
>Arrival-Date: Sun Nov 30 09:32:01 EST 1997
>Last-Modified:
>Originator: Evan Champion
>Organization:

Show quoted text
>Release: krb5-1.0.3
>Environment:

System: FreeBSD piano.synapse.net 3.0-CURRENT FreeBSD 3.0-CURRENT #0: Thu Nov 27 09:25:55 EST 1997 evanc@piano.synapse.net:/usr/src/sys/compile/PIANO i386


Show quoted text
>Description:
FreeBSD already has a libcom_err.a (distributed standard in
/usr/lib). The code is almost identical; krb5's version just
has a lot of extra support for Mac's, PC's, etc.

Since the code is already there, it seems a bit wasteful to
compile it again.
Show quoted text
>How-To-Repeat:

Show quoted text
>Fix:
Check to see if system already has libcom_err in standard
library directories (/usr/lib, /usr/local/lib). If so, don't
compile libcom_err again.

Should do the same for libdb if the system one is found
to be version 2.

Perhaps allow an override, ie: --with-krb5-libcom_err,
--with-krb5-libdb.
Show quoted text
>Audit-Trail:
>Unformatted:
Download (untitled) / with headers
text/plain 3.5KiB
From hartmans@MIT.EDU Thu Nov 21 17:32:56 1996
Received: from MIT.EDU (SOUTH-STATION-ANNEX.MIT.EDU [18.72.1.2]) by rt-11.MIT.EDU (8.7.5/8.7.3) with SMTP id RAA05668 for <bugs@RT-11.MIT.EDU>; Thu, 21 Nov 1996 17:32:55 -0500
Received: from PORTNOY.MIT.EDU by MIT.EDU with SMTP
id AA14158; Thu, 21 Nov 96 17:32:51 EST
Received: by portnoy.MIT.EDU (5.x/4.7) id AA21655; Thu, 21 Nov 1996 17:32:49 -0500
Message-Id: <9611212232.AA21655@portnoy.MIT.EDU>
Date: Thu, 21 Nov 1996 17:32:49 -0500
From: hartmans@MIT.EDU
Reply-To: hartmans@MIT.EDU
To: krb5-bugs@MIT.EDU
Cc: ghudson@MIT.EDU, mycroft@MIT.EDU
Subject: krb5 should consider using installed com_err and installing compile_et
X-Send-Pr-Version: 3.99

Show quoted text
>Number: 217
>Category: krb5-build
>Synopsis: krb5 should consider using installed com_err and installing compile_et
>Confidential: no
>Severity: serious
>Priority: low
>Responsible: tlyu
>State: open
>Class: change-request
>Submitter-Id: unknown
>Arrival-Date: Thu Nov 21 17:33:01 EST 1996
>Last-Modified: Tue Nov 26 02:41:00 EST 1996
>Originator: Sam Hartman
>Organization:
mit
Show quoted text
>Release: 1.0-development
>Environment:

System: SunOS portnoy 5.4 Generic_101945-37 sun4m sparc


Show quoted text
>Description:


Greg is planning to fix Zephyr so that it installes
compile_et. He argues that if the library is installed, tools to use
it should be installed. The library should be installed in both krb5
and Zephyr. I think we should coordiante with Zephyr and Athena on this because there is no good reason not to.

Charles would like to be able to use a native libcom_err and
compile_et if one exists. We should probably consider this as it
would be useful for Athena as well as others.



Show quoted text
>How-To-Repeat:

Show quoted text
>Fix:

Show quoted text
>Audit-Trail:

From: "Theodore Y. Ts'o" <tytso@MIT.EDU>
To: krb5-bugs@MIT.EDU, hartmans@MIT.EDU
Cc: tlyu@MIT.EDU, gnats-admin@RT-11.MIT.EDU, krb5-prs@RT-11.MIT.EDU
Subject: Re: krb5-build/217: krb5 should consider using installed com_err and installing compile_et
Date: Tue, 26 Nov 1996 01:54:23 -0500

The problem with installing compile_et is that we don't compile it.
Instead, we use the awk scripts for portability reasons.

From: Sam Hartman <hartmans@MIT.EDU>
To: "Theodore Y. Ts'o" <tytso@MIT.EDU>
Cc: krb5-bugs@MIT.EDU, tlyu@MIT.EDU, gnats-admin@RT-11.MIT.EDU,
krb5-prs@RT-11.MIT.EDU
Subject: Re: krb5-build/217: krb5 should consider using installed com_err and installing compile_et
Date: 26 Nov 1996 02:31:23 -0500

Show quoted text
>>>>> ""Theodore" == "Theodore Y Ts'o" <tytso@MIT.EDU> writes:

"Theodore> The problem with installing compile_et is that we don't
"Theodore> compile it. Instead, we use the awk scripts for
"Theodore> portability reasons.

You can install an executable awk script.

From: "Theodore Y. Ts'o" <tytso@MIT.EDU>
To: Sam Hartman <hartmans@MIT.EDU>
Cc: "Theodore Y. Ts'o" <tytso@MIT.EDU>, krb5-bugs@MIT.EDU, tlyu@MIT.EDU,
gnats-admin@RT-11.MIT.EDU, krb5-prs@RT-11.MIT.EDU
Subject: Re: krb5-build/217: krb5 should consider using installed com_err and installing compile_et
Date: Tue, 26 Nov 1996 02:40:22 -0500

From: Sam Hartman <hartmans@MIT.EDU>
Date: 26 Nov 1996 02:31:23 -0500

Show quoted text
>>>>> ""Theodore" == "Theodore Y Ts'o" <tytso@MIT.EDU> writes:

"Theodore> The problem with installing compile_et is that we don't
"Theodore> compile it. Instead, we use the awk scripts for
"Theodore> portability reasons.

You can install an executable awk script.

Yes, but there are *two* awk scripts....
Show quoted text
>Unformatted:
Subject: krb5 should be able to use a separate com_err
We would like to be able to specify in configure the ability to use a
com_err library and compile_et binary in a specific location rather than
building com_err in the tree.

KfM needs this because it uses its own com_err which supports Mac OS X's
string localization mechanism rather than hard-coded strings.
The changes checked in on the trunk add a new configure option
--with-system-et that will cause krb5 to use installed versions of
compile_et and libcom_err. (And --with-system-ss for mk_cmds and
libss.) No API change has been made; for "normal" (i.e., non-Mac) UNIX,
it is assumed that the same API can be used as before, calling
add/remove_error_table on et_FOO_error_table; doing otherwise would mean
making the library unsafe for dlopen/dlclose.