update configure.ac to autoconf 2.61
PATCH 1/1 update configure.ac to autoconf 2.61
Attached patch is generated via `autoupdate --force' from autoconf 2.61 a svn-diff after ./autogen.sh doestn't show anything more, because the ./configure script in SVN is already generated with autoconf 2.61.
autoconf 2.61 is dated 17.11.2006 on ftp.gnu.org
Debian etch + lenny have autoconf 2.61
etch has been released 8th April 2007
Ubuntu has it since feisty which has been released 19th April 2007
I haven't checked the others but I think this is recent enough for the configure.ac (remember it's generated by ./autogen.sh anyway)
Yes I really want to get rid of old stuff and if you even only need to call `autoupdate --force' ;)
Debian experimental has autoconf 2.62 which I normally use, there is some other stuff deprecated and so changed by `autoupdate' but I'll leave that for the future.
Signature · 1 line
Felix Zielcke
2008-09-03 Felix Zielcke <[email protected]>
* configure.ac (AC_INIT): Quote version number. (AC_PREREQ): Bumped to 2.61. (AC_TRY_COMPILE): Replace deprecated macro with .. (AC_COMPILE_IFELSE): ... this.
configure.ac +4 -4
@@ -31,8 +31,8 @@ dnl (such as TARGET_CC, TARGET_CFLAGS, e dnl type. -AC_INIT(GRUB, 1.96, [[email protected]]) -AC_PREREQ(2.53) +AC_INIT([GRUB],[1.96],[[email protected]]) +AC_PREREQ(2.61) AC_CONFIG_SRCDIR([include/grub/dl.h]) AC_CONFIG_HEADER([config.h])
@@ -266,7 +266,7 @@ if test "x$TARGET_CFLAGS" = x; then # optimization flags. AC_CACHE_CHECK([whether optimization for size works], grub_cv_cc_Os, [ CFLAGS=-Os - AC_TRY_COMPILE(, , grub_cv_cc_Os=yes, grub_cv_cc_Os=no) + AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[]], [[]])],[grub_cv_cc_Os=yes],[grub_cv_cc_Os=no]) ]) if test "x$grub_cv_cc_Os" = xyes; then TARGET_CFLAGS="$TARGET_CFLAGS -Os"
@@ -278,7 +278,7 @@ if test "x$TARGET_CFLAGS" = x; then if test "x$target_cpu" = xi386; then AC_CACHE_CHECK([whether -falign-loops works], [grub_cv_cc_falign_loop], [ CFLAGS="-falign-loops=1" - AC_TRY_COMPILE(, , [grub_cv_cc_falign_loop=yes], [grub_cv_cc_falign_loop=no]) + AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[]], [[]])],[grub_cv_cc_falign_loop=yes],[grub_cv_cc_falign_loop=no]) ]) if test "x$grub_cv_cc_falign_loop" = xyes; then
Am Mittwoch, den 03.09.2008, 15:27 +0200 schrieb Felix Zielcke:
Quoted 3 lines
Yes I really want to get rid of old stuff and if you even only need to call `autoupdate --force' ;)
Oh and config.guess and config.sub have
timestamp='2007-07-22'
whereas the versions from automake 1.9 and 1.10 have timestamp='2008-01-23
On a Debian system you find them in /usr/share/automake-<version>/
Any objection that I update these 2 files too?
Attached is a patch for these 2 files.
Not really necessary but well a bit outdated ;)
Signature · 1 line
Felix Zielcke
2008-09-03 Felix Zielcke <[email protected]>
* config.guess: Update to version from `2008-01-23'
* config.sub: Likewise.
config.guess +21 -11
@@ -1,10 +1,10 @@ #! /bin/sh # Attempt to guess a canonical system name. # Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, -# 2000, 2001, 2002, 2003, 2004, 2005, 2006 Free Software Foundation, -# Inc. +# 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008 +# Free Software Foundation, Inc. -timestamp='2007-07-22' +timestamp='2008-01-23' # This file is free software; you can redistribute it and/or modify it # under the terms of the GNU General Public License as published by
@@ -56,8 +56,8 @@ version="\ GNU config.guess ($timestamp) Originally written by Per Bothner. -Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005 -Free Software Foundation, Inc. +Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, +2002, 2003, 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE."
@@ -532,7 +532,7 @@ EOF echo rs6000-ibm-aix3.2 fi exit ;; - *:AIX:*:[45]) + *:AIX:*:[456]) IBM_CPU_ID=`/usr/sbin/lsdev -C -c processor -S available | sed 1q | awk '{ print $1 }'` if /usr/sbin/lsattr -El ${IBM_CPU_ID} | grep ' POWER' >/dev/null 2>&1; then IBM_ARCH=rs6000
@@ -799,6 +799,9 @@ EOF EM64T | authenticamd) echo x86_64-unknown-interix${UNAME_RELEASE} exit ;; + IA64) + echo ia64-unknown-interix${UNAME_RELEASE} + exit ;; esac ;; [345]86:Windows_95:* | [345]86:Windows_98:* | [345]86:Windows_NT:*) echo i${UNAME_MACHINE}-pc-mks
@@ -833,7 +836,14 @@ EOF echo ${UNAME_MACHINE}-pc-minix exit ;; arm*:Linux:*:*) - echo ${UNAME_MACHINE}-unknown-linux-gnu + eval $set_cc_for_build + if echo __ARM_EABI__ | $CC_FOR_BUILD -E - 2>/dev/null \ + | grep -q __ARM_EABI__ + then + echo ${UNAME_MACHINE}-unknown-linux-gnu + else + echo ${UNAME_MACHINE}-unknown-linux-gnueabi + fi exit ;; avr32*:Linux:*:*) echo ${UNAME_MACHINE}-unknown-linux-gnu
@@ -954,8 +964,8 @@ EOF x86_64:Linux:*:*) echo x86_64-unknown-linux-gnu exit ;; - xtensa:Linux:*:*) - echo xtensa-unknown-linux-gnu + xtensa*:Linux:*:*) + echo ${UNAME_MACHINE}-unknown-linux-gnu exit ;; i*86:Linux:*:*) # The BFD linker knows what the default object file format is, so
@@ -1474,9 +1484,9 @@ This script, last modified $timestamp, h the operating system you are using. It is advised that you download the most up to date version of the config scripts from - http://savannah.gnu.org/cgi-bin/viewcvs/*checkout*/config/config/config.guess + http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.guess;hb=HEAD and - http://savannah.gnu.org/cgi-bin/viewcvs/*checkout*/config/config/config.sub + http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.sub;hb=HEAD If the version you run ($0) is already up to date, please send the following data and any information you think might be
config.sub +38 -6
90 lines not on this page · raw message
Quoted 10 lines · Felix Zielcke
Am Mittwoch, den 03.09.2008, 15:27 +0200 schrieb Felix Zielcke:
Quoted 3 lines
Yes I really want to get rid of old stuff and if you even only need to call `autoupdate --force' ;)
Oh and config.guess and config.sub have
timestamp='2007-07-22'whereas the versions from automake 1.9 and 1.10 have timestamp='2008-01-23
You can get the current config.* files from CVS:
cvs -d :pserver:[email protected]:/sources/config co config
They are maintained in a very conservative matter, so it may be better to use the CVS versions. Actually, it's possible that they are the same as those you have.
Quoted 6 lines
On a Debian system you find them in /usr/share/automake-<version>/
Any objection that I update these 2 files too?
Attached is a patch for these 2 files.Not really necessary but well a bit outdated ;)
Fine with me.
Signature · 2 lines
Regards, Pavel Roskin
Am Mittwoch, den 03.09.2008, 19:47 -0400 schrieb Pavel Roskin:
Quoted 6 lines
You can get the current config.* files from CVS:
cvs -d :pserver:[email protected]:/sources/config co configThey are maintained in a very conservative matter, so it may be better to use the CVS versions. Actually, it's possible that they are the same as those you have.
They use now git as you can even see in the patch from my mail but I was too lazy to check because I thought that automake-1.10 would have a recent version.
But it seems not :)
Quoted 1 line
Fine with me.
Commited these 2 files.
Hurray a bit less outdated stuff now :)
Signature · 1 line
Felix Zielcke
Quoted 10 lines · Felix Zielcke
Attached patch is generated via `autoupdate --force' from autoconf 2.61 a svn-diff after ./autogen.sh doestn't show anything more, because the ./configure script in SVN is already generated with autoconf 2.61.
autoconf 2.61 is dated 17.11.2006 on ftp.gnu.org
Debian etch + lenny have autoconf 2.61
etch has been released 8th April 2007
Ubuntu has it since feisty which has been released 19th April 2007
Quoted 5 lines
I haven't checked the others but I think this is recent enough for the configure.ac (remember it's generated by ./autogen.sh anyway)
Yes I really want to get rid of old stuff and if you even only need to call `autoupdate --force' ;)
I don't see any old stuff being removed. I don't see any justification for any of the changes. The new code is longer and less readable. I don't think AC_TRY_COMPILE is a big problem yet. It's not like it won't produce a valid test.
In any case, please try to avoid extra long lines where it can be easily avoided.
Okuji wrote that he uses Autoconf 2.59, so it would be nice to check that your changes would still work with that version.
Signature · 2 lines
Regards, Pavel Roskin
Am Mittwoch, den 03.09.2008, 19:43 -0400 schrieb Pavel Roskin:
Quoted 5 lines
I don't see any old stuff being removed. I don't see any justification for any of the changes. The new code is longer and less readable. I don't think
AC_TRY_COMPILEis a big problem yet. It's not like it won't produce a valid test.
http://www.gnu.org/software/autoconf/manual/autoconf.html#Autoconf-Macro-Index
AC_TRY_COMPILE is listed as `Obsolete Macro'.
I attached now even a diff with autoconf 2.62 which I won't commit, but maybe you or someone else is interested.
Changelog says AC_GNU_SOURCE has been deprecated, though they're currently no plans to remove it.
Currently it's not really needed but things get easily forgotten, even I myself forgot a patch from me ..
We don't need to discuss running `autoupdate' in ./autogen.sh this is really bad.
The Debian Build System has such a feature too but ftp-masters' reject packages which uses it by default.
You only need to run autoconf -Wall then you see.
Ok they're not shown by default, but I don't like to have obsolte deprecated stuff if it can be easily avoided like in this case.
Quoted 2 lines
In any case, please try to avoid extra long lines where it can be easily avoided.
I hope the intention is now right, Emacs had a problem with it.
Quoted 2 lines
Okuji wrote that he uses Autoconf 2.59, so it would be nice to check that your changes would still work with that version.
Luckly it's still avaible in Debian oldstable (sarge) newest CentOS 5.2 even still has autoconf 2.59 Changes are the same as with 2.61 so 2.59 seems to be fine.
Signature · 1 line
Felix Zielcke
configure.ac +5 -5
@@ -31,8 +31,8 @@ dnl (such as TARGET_CC, TARGET_CFLAGS, e dnl type. -AC_INIT(GRUB, 1.96, [[email protected]]) -AC_PREREQ(2.53) +AC_INIT([GRUB],[1.96],[[email protected]]) +AC_PREREQ(2.62) AC_CONFIG_SRCDIR([include/grub/dl.h]) AC_CONFIG_HEADER([config.h])
@@ -150,7 +150,7 @@ AC_PROG_CC # Must be GCC. test "x$GCC" = xyes || AC_MSG_ERROR([GCC is required]) -AC_GNU_SOURCE +AC_USE_SYSTEM_EXTENSIONS AC_SYS_LARGEFILE # Identify characteristics of the host architecture.
@@ -266,7 +266,7 @@ if test "x$TARGET_CFLAGS" = x; then # optimization flags. AC_CACHE_CHECK([whether optimization for size works], grub_cv_cc_Os, [ CFLAGS=-Os - AC_TRY_COMPILE(, , grub_cv_cc_Os=yes, grub_cv_cc_Os=no) + AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[]], [[]])],[grub_cv_cc_Os=yes],[grub_cv_cc_Os=no]) ]) if test "x$grub_cv_cc_Os" = xyes; then TARGET_CFLAGS="$TARGET_CFLAGS -Os"
@@ -278,7 +278,7 @@ if test "x$TARGET_CFLAGS" = x; then if test "x$target_cpu" = xi386; then AC_CACHE_CHECK([whether -falign-loops works], [grub_cv_cc_falign_loop], [ CFLAGS="-falign-loops=1" - AC_TRY_COMPILE(, , [grub_cv_cc_falign_loop=yes], [grub_cv_cc_falign_loop=no]) + AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[]], [[]])],[grub_cv_cc_falign_loop=yes],[grub_cv_cc_falign_loop=no]) ]) if test "x$grub_cv_cc_falign_loop" = xyes; then
config.h.in +36 -5
53 lines not on this page · raw message
aclocal.m4 +27 -27
104 lines not on this page · raw message
configure.ac +8 -4
@@ -31,8 +31,8 @@ dnl (such as TARGET_CC, TARGET_CFLAGS, e dnl type. -AC_INIT(GRUB, 1.96, [[email protected]]) -AC_PREREQ(2.53) +AC_INIT([GRUB],[1.96],[[email protected]]) +AC_PREREQ(2.59) AC_CONFIG_SRCDIR([include/grub/dl.h]) AC_CONFIG_HEADER([config.h])
@@ -266,7 +266,9 @@ if test "x$TARGET_CFLAGS" = x; then # optimization flags. AC_CACHE_CHECK([whether optimization for size works], grub_cv_cc_Os, [ CFLAGS=-Os - AC_TRY_COMPILE(, , grub_cv_cc_Os=yes, grub_cv_cc_Os=no) + AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[]], [[]])], + [grub_cv_cc_Os=yes], + [grub_cv_cc_Os=no]) ]) if test "x$grub_cv_cc_Os" = xyes; then TARGET_CFLAGS="$TARGET_CFLAGS -Os"
@@ -278,7 +280,9 @@ if test "x$TARGET_CFLAGS" = x; then if test "x$target_cpu" = xi386; then AC_CACHE_CHECK([whether -falign-loops works], [grub_cv_cc_falign_loop], [ CFLAGS="-falign-loops=1" - AC_TRY_COMPILE(, , [grub_cv_cc_falign_loop=yes], [grub_cv_cc_falign_loop=no]) + AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[]], [[]])], + [grub_cv_cc_falign_loop=yes], + [grub_cv_cc_falign_loop=no]) ]) if test "x$grub_cv_cc_falign_loop" = xyes; then
Quoted 6 lines · Felix Zielcke
Quoted 2 lines
Okuji wrote that he uses Autoconf 2.59, so it would be nice to check that your changes would still work with that version.
Luckly it's still avaible in Debian oldstable (sarge) newest CentOS 5.2 even still has autoconf 2.59 Changes are the same as with 2.61 so 2.59 seems to be fine.
I'm fine with the shorter patch, but I'd like you to wait a couple of days in case there are any objections.
By the way, config.h.in is a generated file, so changes to it don't need to be published. The same applies to configure and *.mk files.
Signature · 2 lines
Regards, Pavel Roskin
Am Donnerstag, den 04.09.2008, 12:49 -0400 schrieb Pavel Roskin:
Quoted 2 lines
I'm fine with the shorter patch, but I'd like you to wait a couple of days in case there are any objections.
Vesa said already on IRC to me that it's fine for me to but he suggests waiting vor Okuji or Marco which I wanted to do anyway (at least with the 2.61 requirement)
Now 2.59 shouldn't be any problem at all but yeah just in case :)
Quoted 2 lines
By the way, config.h.in is a generated file, so changes to it don't need to be published. The same applies to configure and *.mk files.
Yep Autoheader generates it, I just wanted to show the difference with the newer `AC_USE_SYSTEM_EXTENSIONS' in case someone is interested :) I saw it now even in the docs it's only supported with autoconf >= 2.60
and because I just don't know enough if it would be safe for us to just start using it I downgraded autoconf to 2.61 for `autupdate'.
It would be better if we use for the SVN ./configure only one autoconf version.
I just noticed that with current SVN and a ./autogen.sh with autoconf 2.62 creates the config.h.in in my 2.62 patch, which is the only difference.
So this means that AC_GNU_SOURCE will be in the future just an alias to it.
A little thing we maybe shouldn't forget totally :)
But for now this is really too early to change configure.ac to it.
Signature · 1 line
Felix Zielcke
Well, commited.
Am Donnerstag, den 04.09.2008, 09:45 +0200 schrieb Felix Zielcke:
Quoted 44 lines
Am Mittwoch, den 03.09.2008, 19:43 -0400 schrieb Pavel Roskin:
Quoted 5 lines
I don't see any old stuff being removed. I don't see any justification for any of the changes. The new code is longer and less readable. I don't think
AC_TRY_COMPILEis a big problem yet. It's not like it won't produce a valid test.http://www.gnu.org/software/autoconf/manual/autoconf.html#Autoconf-Macro-Index
AC_TRY_COMPILEis listed as `Obsolete Macro'.I attached now even a diff with autoconf 2.62 which I won't commit, but maybe you or someone else is interested.
Changelog saysAC_GNU_SOURCEhas been deprecated, though they're currently no plans to remove it.Currently it's not really needed but things get easily forgotten, even I myself forgot a patch from me ..
We don't need to discuss running `autoupdate' in ./autogen.sh this is really bad.
The Debian Build System has such a feature too but ftp-masters' reject packages which uses it by default.You only need to run autoconf -Wall then you see.
Ok they're not shown by default, but I don't like to have obsolte deprecated stuff if it can be easily avoided like in this case.Quoted 2 lines
In any case, please try to avoid extra long lines where it can be easily avoided.
I hope the intention is now right, Emacs had a problem with it.
Quoted 2 lines
Okuji wrote that he uses Autoconf 2.59, so it would be nice to check that your changes would still work with that version.
Luckly it's still avaible in Debian oldstable (sarge) newest CentOS 5.2 even still has autoconf 2.59 Changes are the same as with 2.61 so 2.59 seems to be fine.
_______________________________________________
Grub-devel mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/grub-devel