From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx1.redhat.com ([209.132.183.28]:57020 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755952Ab0JLB3z (ORCPT ); Mon, 11 Oct 2010 21:29:55 -0400 From: Jeff Layton To: steved@redhat.com Cc: chuck.lever@oracle.com, linux-nfs@vger.kernel.org Subject: [PATCH] nfs-utils: fix default value for --enable-tirpc Date: Mon, 11 Oct 2010 21:29:51 -0400 Message-Id: <1286846991-4876-1-git-send-email-jlayton@redhat.com> Sender: linux-nfs-owner@vger.kernel.org List-ID: Content-Type: text/plain MIME-Version: 1.0 The patch I originally proposed to make --enable-tirpc the default had the default for $enable_tirpc as a blank string. When it was merged however, that value was changed to a 'yes'. This broke the ability for autoconf to detect and autodisable building with tirpc. When tirpc isn't present now, the configure step fails unless someone explicitly specifies --disable-libtirpc. We need $enable_tirpc to be a tristate. 'yes' means that someone explicitly requested building with tirpc. 'no' means that it was explicitly disabled. Anything else means that no one specified a value. Fix it by setting the value to a blank string so that the default is properly undefined. Reported-by: Chuck Lever Signed-off-by: Jeff Layton --- configure.ac | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/configure.ac b/configure.ac index 3058be6..5408e85 100644 --- a/configure.ac +++ b/configure.ac @@ -136,7 +136,7 @@ AC_ARG_ENABLE(tirpc, [AC_HELP_STRING([--enable-tirpc], [enable use of TI-RPC @<:@default=yes@:>@])], enable_tirpc=$enableval, - enable_tirpc='yes') + enable_tirpc='') AC_ARG_ENABLE(ipv6, [AC_HELP_STRING([--enable-ipv6], [enable support for IPv6 @<:@default=no@:>@])], -- 1.7.1