the feature currently
offered by autoconf :-(
The unfortunate side-effect of your proposal is that these tests would
suddenly start working, *unless* someone used a cache file. It is
autoconf that has to be extended to support this kind of test. I.e.,
the caching mechanism is not the culprit, i
On Feb 25, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
>>>>>> "Alexandre" == Alexandre Oliva <[EMAIL PROTECTED]> writes:
Alexandre> Like:
Alexandre> How about ``remove config.cache and re-run configure'', as
Alexandre> printed by
f course! Stupid me! :-)
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oliva@{lsd.ic.unicamp.br, gnu.org} Write to mailing lists, not to me
pes.h in their system.h
to fail to find those types at build time.
Maybe we should have a specialized AC_CHECK_HEADER for inttypes.h,
that will add this header to the list of default headers.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat company
C_CHECK_HEADER for
Alexandre> inttypes.h
> Eerk!
:-)
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oliva@{lsd.ic.unicamp
dency tracking
for free, as long as you use CVS automake.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oliva@{lsd.ic.unicamp.br
On Feb 25, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
> How about leaving an implicit test in Autoconf 2.15's configure, with
> a message such as `Please warn us that your machine does not support
> blah blah.
I like this idea
--
Alexandre Oliva http://www.
that should be a fatal error right there, too.
I agree. Any previous tests for libraries, header files, etc, using
one compiler, would be invalid for the other compiler, so it's better
to fail right away. At least before we have explicit support for
mixed cross/non-cross builds.
--
Al
zero would be cheating too much. Do you have any
example of such a cheating compiler?
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD stude
one with regular expressions,
but I heard perl has an extended regexp engine that can do stuff like
that. Unless we won't support arguments containing brackets, in which
case we don't really need this extended support, and regular shell
tools would do.
--
Alexandre Oliva http://www.ic.un
ds just for
configure to abort later on because the compiler doesn't satisfy the
checked needs.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD s
That's probably the one you
should use.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oliva@{lsd.ic.unicamp.br, gnu.org} Write to mailing lists, not to me
take the file name and the
line number as arguments $1 and $3, respectively, I don't think we
should expose this to the users. $1 should be the first actual
argument to the macro. We should probably do the translation
internally.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva
nsion, that, in the example above, would be defined as:
define([AC_SUBST], [some-junk $1 some-garbage])
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and Evangeli
On Feb 15, 2000, Ossama Othman <[EMAIL PROTECTED]> wrote:
> Hi Alexandre,
> On Tue, Feb 15, 2000 at 11:10:40AM -0200, Alexandre Oliva wrote:
>> On Feb 12, 2000, Ossama Othman <[EMAIL PROTECTED]> wrote:
>>
>> > I think it's bad idea to check for bool
reprocessor. And there might be C++
compilers that do not define __cplusplus.
Let's use the test you (?) had suggested, at least as a default test.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
F
ng
> to be aware of.
Libtool has a rather elaborate autoconf macro to check for a
BSD-compatible working nm and to create a sed script to extract symbol
names.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, c
the maintainer of
config.{guess,sub}, at [EMAIL PROTECTED]
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oliva@{lsd.ic.unicamp
([option-name is deprecated, use alternate-option instead])
fi
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oliva@{lsd.ic.unic
gt; of detailed mode, reporting cpu types. What do you think, would
> that be doable?
Doesn't it do it already? I mean, it says i[3456]86,
alpha{,ev[5678]*}, sparc{,64,v9}, etc. Just not for all systems.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus
; misleading to name it AM_.
I disagree. In fact, I often recommend that people start their own
macros with AC_uid_, so that autoconf prints an error if the macro
fails to be properly expanded in the configure script.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygn
x27;s
> signed. Revised patch attached.
But it may fail to compile if char is unsigned and
sizeof(char)==sizeof(unsigned int).
How about:
switch (c) { case 0: case (((char)-1) >= 0): ; }
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat c
On Feb 23, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
>>>>>> "Alexandre" == Alexandre Oliva <[EMAIL PROTECTED]> writes:
Alexandre> I disagree. In fact, I often recommend that people start
Alexandre> their own macros with AC_uid_, so that au
be nice to have this kind of information printed by
config.guess. But there may be backward compatibility issues. Ben
will know what to do.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Softw
On Feb 23, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
>>>>>> "Alexandre" == Alexandre Oliva <[EMAIL PROTECTED]> writes:
Alexandre> On Feb 23, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
>>>>>>>> "Alexandr
m/autoconf/ .
It's not a problem. The CVS tree has moved to
:pserver:[EMAIL PROTECTED]:/home/cvs module autoconf. That
was even announced in this mailing list! :-)
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyao
erver:[EMAIL PROTECTED], module autoconf.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oliva@{lsd.ic.unicamp.br, gnu.org} Writ
ons too, module config.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oliva@{lsd.ic.unicamp.br, gnu.org} Write to mailing lists, not to me
even when config.guess couldn't find a compiler,
it still guessed as much as possible from the unames, even if that
won't enable processor-specific optimizations. For example, if you
can't compile a x86 or alpha test to detect i686 or alphaev8, just say
i386 or alpha.
--
Alexandre
fine. It's just that, in the autoconf docs and implementation,
we've decided to consistently use #if instead of #ifdef.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer
a literal ^Q in it, which you need if you want
> this to work properly. BTW, is it normal that bash does not split
> around ^A? I found nothing like this in the doc:
I don't think it's good practice to embed such weird characters in
shell scripts. I'd rather expand the tr
On Mar 8, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
> I'm OK with this patch.
Ditto.
> One yes is still expected.
You got it :-)
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat,
On Mar 8, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
> I suggest to simply remove the second inclusion of unistd.h.
Agreed.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free
On Mar 8, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
> Now the question is, do we want to pollute
> autoconf.sh, a maintainer tool, with portability issues?
Yes, at least in this case.
> If yes, we need to
> eval set dummy $list
> shift
So be it. It's sim
elt in ${1+"$@"}; do
Some shells will expand "$@" to "" even if $# = 0.
But then, this doesn't apply here, since we know we've got at least
one argument.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red
hat's the point). Anyway, I was wondering if we
couldn't do it some other way, For example, by adding a line-break
after $1, and editing it out somehow after the whole thing is
processed:
define([m4_noquote],[m4_remove_trailing_newline([changequote(,)$1
changequote([,])])])
--
On Mar 3, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
> | [/\\*] | [A-z]:[/\\*]
> You probably mean
> [/\\]* | [A-z]:[/\\]* )
Actually
[/\\]* | [A-Za-z]:[/\\]* )
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat compan
r code could contain other tests, that could then
assume conftest* don't exist, or even `exit' statements, that would
leave conftest* files behind.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.c
re esthetic means to achieve this.
How about simply not defining it at all? People can always AC_DEFINE
or AC_SUBST them on their own.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software De
user have to specify all sub-packages,
like PACKAGE1,PACKAGE2:--foo?
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oliva@{lsd.i
On Mar 10, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
>>>>>> "Alexandre" == Alexandre Oliva <[EMAIL PROTECTED]> writes:
Alexandre> How about simply not defining it at all? People can always
Alexandre> AC_DEFINE or AC_SUBST them on thei
"
for e in \
ARG_VAR \
CHECK_"(DECL|FILE|FUNC|HEADER|MEMBER|PROG|SIZEOF|TYPE)S?" \
...
do
expr="$expr|^AC_$e\$"
done
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
F
ays use two egreps, one that will match
'^AC_(...)$' and another that matches the remaining alternatives.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and Evange
On Mar 10, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
> Hey, I have an obscure idea :-]
^ :-)
> How about using m4 to filter out those guys :)
Yep, I had thought of that. But if we can make egrep work, m4 is
overkill.
--
Alexandre Ol
On Mar 10, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
> This might be useful for some: better support of Autoconf in C-x 4 a:
Great! Make sure you send this to the maintainers of GNU Emacs and
XEmacs.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guara
e top of ChangeLog, instead of letting patch try to find
> the place thanks to context (which is usually wrong).
Try cvsdiff, clcleanup, cl2patch and clcommit in cvs-utils, at
http://www.ic.unicamp.br/~oliva/snapshots/cvs-utils
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Gua
Makefile.in, so it's fine with me.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oliva@{lsd.ic.unicamp.br, gnu.org} Write to mailing lists, not to me
Putting "nice -n 20 myprogram" is not portable.
> Well, I got no reply, so here is my solution:
Would you mind submitting it to the autoconf macro archive?
http://peti.cys.de/autoconf-archive/
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutio
ctory and create another one under their control, but that's not
something we should be expected to work around in shell-scripts;
having /tmp with the sticky bit enabled takes care of this problem,
and most systems are configured as such.
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva
ortable to let the `#' be at a different
> column than the first.
Yep. That's the work-around for `#error' to work with K&R
preprocessors: just insert whitespace before it. But using `Choke
me.' is better practice for autoconf.
--
Alexandre Oliva http://www
love this tortured way to do simple things. Just give them a
> couple a weeks, and the Debian guys will come up requiring that we use
> mktemp :)
Good idea :-) :-)
(indeed, this would be perfect way to deal with this problem, but
mktemp is not portable itself, so we'd always need a fal
e bet, not KnR?
I'm not sure it's not KnR, but I'm pretty sure there are compilers
that don't support it.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Develope
t; Since autoconf is for portibility you can't really assume that /tmp
> exists.
If /tmp doesn't exist, pretty much everything breaks, unless TMPDIR is
set. And it does use $TMPDIR, so we're fine.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
ave object files in sub-directories, and that is nowhere
advised that he must also upgrade autoconf, otherwise he'll get weird
error messages from [s]hell! :-)
--
Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Enjoy Guaraná
Cygnus Solutions, a Red Hat companyaoliva@{redhat,
nge for `autoconf' to preserve that
section, and we'd need some tool to add/remove/update shtools within
that special section of configure. Comments?
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, c
On Mar 13, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
>>>>>> "Alexandre" == Alexandre Oliva <[EMAIL PROTECTED]> writes:
Alexandre> On Mar 13, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
>>> And again, I shall emphasize that this l
On Feb 2, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
>>>>>> "Alexandre" == Alexandre Oliva <[EMAIL PROTECTED]> writes:
Alexandre> Only CVS autoconf accepts implicitly quotes backticks in
Alexandre> MSGs. But it complains if backtick is alread
ors. We should probably try
to find some way that does the right thing even in such contexts.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD st
On Mar 18, 2000, Felix Lee <[EMAIL PROTECTED]> wrote:
> Alexandre Oliva <[EMAIL PROTECTED]>:
>> BTW, this won't work as expected if `configure' is generated on
>> platforms that use CR+LF as line separators. We should probably try
>> to find some
On Mar 19, 2000, Felix Lee <[EMAIL PROTECTED]> wrote:
> Alexandre Oliva <[EMAIL PROTECTED]>:
>> This assumes ASCII. We've just removed the latest assumptions of
>> ASCII, so it wouldn't be wise to re-introduce them :-)
> oh. hmm. then how about using
'm not really sure `\t' and `\n' are portable AWK.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oliva@{lsd.ic.unic
close the fd to foo before the mv/rm.
IIRC, this had already been fixed in autoheader after release 2.13.
Was the problem re-introduced?
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Softwar
On Mar 20, 2000, Earnie Boyd <[EMAIL PROTECTED]> wrote:
> Ah. That's nice to know. I have release 2.13. I need to find a
> good free ssh client for Win32 so that I can download the CVS
> version.
You don't need ssh for :pserver:anoncvs
> Any suggestions?
Nope,
led. Please post
it, along with the patch, to [EMAIL PROTECTED] Thanks!
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oliva@{lsd.i
On Mar 27, 2000, Pavel Roskin <[EMAIL PROTECTED]> wrote:
> * Major changes in release 2.14a
> Shouldn't is read:
> * Major changes since release 2.13
How about `major changes since last release'?
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~o
;s just that, if we write it in the NEWS file before thea
actual release, check-dist won't remind us of updating NEWS before a
release.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Softwa
On Mar 27, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
>>>>>> "Alexandre" == Alexandre Oliva <[EMAIL PROTECTED]> writes:
>>> What's wring with `Autoconf 2.14a'?
Alexandre> Nothing, it's just that, if we write it in the NEWS
On Mar 27, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
> So I really want to keep it.
Ok
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS Ph
`configure'. And automake could do it automatically for Makefiles.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oliva@{lsd.
amp;& mkdir $$).
> That sounds quite reasonable to me.
I have a feeling that using directories may be slightly safer or less
safe, but I can't seem to decide which one, so I think I'm ok with it
too ;-)
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cyg
\/]', which does work, even with bash 2.04.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oliva@{lsd.ic.unicamp.br, gnu.org}
.in, so that libtool will get configured to use the C++
compiler/linker.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oliva@{l
t MAKE_TRAP as you suggested, and then use
$(MAKE_TRAP) for recursive invocations.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oli
On Mar 31, 2000, "Lars J. Aas" <[EMAIL PROTECTED]> wrote:
> How do I override the rule for entering one of them?
I don't think you can override it :-(
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat company
make like described below
> will fail on some platform?
The trap is fine, but the `:=' syntax is far from portable.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer
On Apr 2, 2000, Erez Zadok <[EMAIL PROTECTED]> wrote:
> Update: I managed to solve that problem by rerunning automake on the .am
> files in the autoconf CVS repository.
You must re-run aclocal before automake.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~o
in/sh
/path/to/missing --run tar"'
However, in my case, the srcdir pathname is absolute. This may make a
difference.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software De
as kind enough to supply us with the fix, but we
need not bother creating such ChangeLog entries ourselves.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS
it append the "-Olimit 2000" option only if the compilation fails
> without the option and succeeds with the option.
Which means we could have a macro to encapsulate all this
functionality :-)
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions
t; I'm ready to do that.
And it keeps getting better :-) :-)
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oliva@{lsd.ic.un
s $RANDOM support, and shouldn't be any worse on hosts
> without it.
Except that RANDOM may be not that RANDOM, i.e., it may randomly clash
too, so we'd better retry it a few times before giving up. It may
become extremely annoying in case of a huge package with dozens of
`configur
:-)
> not really - the given value of -Olimit varies according to the compiler,
> the program that you are trying to compile, and the optimization level.
Actually, I was just thinking of the ``test, if it fails add the flag
and test again, the revert if it still fail''.
--
Ale
On Apr 5, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
>>>>>> "Alexandre" == Alexandre Oliva <[EMAIL PROTECTED]> writes:
Alexandre> Except that RANDOM may be not that RANDOM, i.e., it may
Alexandre> randomly clash too, so we'd better retry
On Apr 7, 2000, "Clere, Jean-Frederic" <[EMAIL PROTECTED]> wrote:
> The config.guess and config.sub have to be modified to add
> this machine so that autoconf works correctly.
Thanks, but these files are now maintained by other people. Please
send patches to [EMAIL PRO
(i.e., terminally strict).
Sounds good, but please post to [EMAIL PROTECTED] and include a
ChangeLog entry.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and Eva
onto that?
Seems like a good idea, as long as you replace --target for --host.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oli
On Apr 21, 2000, Mo DeJong <[EMAIL PROTECTED]> wrote:
> Ok, I can buy that --host argument, but then why does using
> --host also not work?
Because that was never implemented?
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red
> Should autoconf know about the host it is running under and the
> cross "host" info?
Yep. That's what AC_CANONICAL_BUILD is for.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.
ut the target platform when building the
toolchain.
> No that you mention it, the 2.13 docs do not even mention
> AC_CANONICAL_HOST. Was this something that was added recently?
No, it's not recent, but I think it was supposed to be something
internal.
--
Alexandre OlivaEnjoy Guar
so that cross compiling works properly.
Sounds good to me (i.e., you've got one approval; a patch needs two to
be installed in the autoconf CVS tree).
Next time, please post it to [EMAIL PROTECTED]
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus So
r the right compiler
> but it does not actually use it.
Is CXX set in your environment?
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-
On Apr 21, 2000, Mo DeJong <[EMAIL PROTECTED]> wrote:
> On 21 Apr 2000, Alexandre Oliva wrote:
>> On Apr 21, 2000, Mo DeJong <[EMAIL PROTECTED]> wrote:
>>
>> > So if it should have been "internal" then why do I need to add the
>> > AC_CAN
ane approach, because the build platform is
the only thing we can actually guess. And then, if you specify a
host, it should affect the default for target, but not for build.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@
On Apr 21, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
> To make my question about $nonopt clearer, I'd like to know if there
> is any reason to distinguish HOST when it comes from --host=HOST, and
> HOST went it comes from ./configure HOST.
I think these should be equiva
re be a strong opposition to its complete removal?
Certainly.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oliva@{lsd.ic.unicamp.br, gnu.org} Write to mailing lists, not to me
On Apr 25, 2000, Lars Hecking <[EMAIL PROTECTED]> wrote:
> One thing I can't seem to figure out: if acconfig.h exists, is
> autoheader supposed to add all its contents to config.h.in?
Yup
> And if this doesn't happen, is there a way to find out why?
Run it with `sh
On Apr 25, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
>>>>>> "Alexandre" == Alexandre Oliva <[EMAIL PROTECTED]> writes:
Alexandre> On Apr 25, 2000, Lars Hecking <[EMAIL PROTECTED]> wrote:
>>> One thing I can't seem to figure o
as equivalent to `./configure
--host=TRIPLET'. I don't see any reason to do it. I'd rather keep
both `--host=TRIPLET' and plain `TRIPLET'.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat
it's not cross-compiling.
--
Alexandre OlivaEnjoy Guaraná, see http://www.ic.unicamp.br/~oliva/
Cygnus Solutions, a Red Hat companyaoliva@{redhat, cygnus}.com
Free Software Developer and EvangelistCS PhD student at IC-Unicamp
oliva@{lsd.ic.unicamp.br, gnu.org} Write to mailing lists, not to me
On Apr 25, 2000, Ian Lance Taylor <[EMAIL PROTECTED]> wrote:
> From: Alexandre Oliva <[EMAIL PROTECTED]>
>Date: 25 Apr 2000 18:31:45 -0300
>On Apr 25, 2000, Ian Lance Taylor <[EMAIL PROTECTED]> wrote:
>> What does --cross add that we don't al
1 - 100 of 531 matches
Mail list logo
