bug-m4
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Bug#154053: sendmailconfig fails due to m4 problems on Linux 2.0.x (fwd)


From: Santiago Vila
Subject: Bug#154053: sendmailconfig fails due to m4 problems on Linux 2.0.x (fwd)
Date: Mon, 29 Jul 2002 18:15:40 +0200 (CEST)

Hello.

I've received the following report from the Debian Bug System.

By looking at some m4 web pages, it seems to be a known problem.
Is there a workaround, other than upgrading to Linux 2.2.x or later?

For example, if the configure script is changed so that it does not
check for sigaltstack and sigstack:

--- configure.orig      Mon Jul 29 18:09:44 2002
+++ configure   Mon Jul 29 18:09:57 2002
@@ -1160,7 +1160,7 @@
 fi


-for ac_func in mkstemp sigaction sigaltstack sigstack sigvec strerror tmpfile
+for ac_func in mkstemp sigaction sigvec strerror tmpfile
 do
 echo $ac_n "checking for $ac_func""... $ac_c" 1>&4
 if eval "test \"`echo '${'ac_cv_func_$ac_func'+set}'`\" = set"; then

would the resulting m4 executable work reliably on every glibc-based system?

[ I know configure.in should be modified instead, but this would
  result in a huge Debian diff for no real gain ].

The report:

---------- Forwarded message ----------
Date: Tue, 23 Jul 2002 17:56:17 -0700
From: David Perlin <address@hidden>
To: Debian Bug Tracking System <address@hidden>
Subject: Bug#154053: sendmailconfig fails due to m4 problems on Linux 2.0.x

Package: sendmail and m4
Version: 8.12.3-4
Severity: grave
Justification: renders package unusable


I was originally doing an upgrade to 3.0 on a system running 2.0.38.
The configuration failed when attempting to build the sendmail config files.
The error report actually seems to come from the M4 preprocessor

m4: sigstack: Function not implemented

I did a little searching around on the web (looked for "sigstack not
found" with Google) and found some interesting discussions about M4 generating
this kind of error when built with 2.2 header files and used on 2.0.x kernels
(sigstack does not exist in 2.0 kernels, and M4 instead of paying attention
to ENOSYS returned by the call and trying to do something else, just fails).
So I was left with a system with no working sendmail config.
I ended up upgrading to a 2.2 kernel to "workaround" the problem.

This is really more of an M4 bug, but it severely affects configuring
sendmail if you upgrade on a system with 2.0 kernels.

Please let me know if I can provide you with any more info.

David Perlin
address@hidden

-- System Information
Debian Release: 3.0
Architecture: i386
Kernel: Linux nanosoft.com 2.0.38 i686
Locale: LANG=C, LC_CTYPE=C




reply via email to

[Prev in Thread] Current Thread [Next in Thread]