[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
AC_LIBTOOL_SYS_MAX_CMD_LEN and CONFIG_SHELL
From: |
Kevin Ryde |
Subject: |
AC_LIBTOOL_SYS_MAX_CMD_LEN and CONFIG_SHELL |
Date: |
Sat, 27 Jul 2002 09:22:03 +1000 |
User-agent: |
Gnus/5.090007 (Oort Gnus v0.07) Emacs/21.1 (i386-debian-linux-gnu) |
In AC_LIBTOOL_SYS_MAX_CMD_LEN I think $CONFIG_SHELL isn't always
available.
Both _LT_AC_PROG_ECHO_BACKSLASH and _AS_LINENO_PREPARE only set it
when they decide to reexec configure, and on ia64 hpux it turns out
that this doesn't happen (due to having "print -r" for echoing, and a
working $LINENO already in /bin/sh already).
The problem only shows up when running "sh configure". If instead "sh
./configure" is used it ends up working because in that case an empty
$CONFIG_SHELL in
$CONFIG_SHELL $0 --fallback-echo "X$testring"
comes out as a normal program run of "./configure --fallback-echo ...".
I think plain $SHELL would be sufficient, and is always set by
autoconf (to a user or reexec supplied $CONFIG_SHELL, or /bin/sh
otherwise).
* config_shell/libtool.m4 (AC_LIBTOOL_SYS_MAX_CMD_LEN): Use $SHELL,
since $CONFIG_SHELL is not always set.
libtool.m4.cmdlen.diff
Description: Source code patch
- AC_LIBTOOL_SYS_MAX_CMD_LEN and CONFIG_SHELL,
Kevin Ryde <=