public inbox for crossgcc@sourceware.org
 help / color / mirror / Atom feed
From: Bryan Hundven <bryanhundven@gmail.com>
To: "Andreas Bießmann" <andreas@biessmann.de>
Cc: Ray Donnelly <mingw.android@gmail.com>,
	"Yann E. MORIN" <yann.morin.1998@free.fr>,
		"crossgcc@sourceware.org" <crossgcc@sourceware.org>
Subject: Re: [PATCH] configure.ac: respect 'g' variants of libtool/libtoolize
Date: Tue, 18 Mar 2014 20:56:00 -0000	[thread overview]
Message-ID: <CAJ+oik14akMHqLSq4sWwnTpeCkezOYZNt+YcD0U9CvV7-T3WsQ@mail.gmail.com> (raw)
In-Reply-To: <CAJ+oik0tHoFs0FqqckX=Rh26UXvukDWDobMNaOrhzd7EqyNiqw@mail.gmail.com>

Andreas, Ray, list,

On Tue, Mar 18, 2014 at 1:54 PM, Bryan Hundven <bryanhundven@gmail.com> wrote:
> Andreas, Ray, list,
>
> On Tue, Mar 18, 2014 at 9:54 AM, Andreas Bießmann <andreas@biessmann.de> wrote:
>> Dear Ray Donnelly,
>>
>> On 18.03.14 17:39, Ray Donnelly wrote:
>>> But using the existing mechanism to specify --with-program= is the way
>>> it is intended to be done, surely?
>>
>> I feel this '--with-program=' switch is used for tools not available in
>> $PATH. For example bootstrapping the _whole_ toolchain beginning with
>> the tools required for ct-ng.
>>
>>> Feel free to disagree of course, and if Yann / others want both
>>> methods then I don't much mind, but you should also implement
>>> detection for the correct vendor and version of the software specified
>>> using the configure options:
>>>
>>>  --with-libtoolize=
>>>  --with-objcopy=
>>>  --with-objdump=
>>>  --with-readelf=
>>>  --with-gperf=
>>>
>>> .. I expect that it might get messy!
>>
>> I don't think so. My patch just provides another default name for the
>> tool in question. In fact it is the correct name where BSD variants and
>> GNU variants collide. This was done before for make and awk too:
>>
>> ---8<---
>> andreas@andreas-mbp % grep AC_PATH_ configure.ac
>>     [AC_PATH_PROGS([$1], [$2])
>>     [AC_PATH_PROGS_FEATURE_CHECK([_BASH], [bash],
>>     [AC_PATH_PROGS_FEATURE_CHECK([_AWK], [awk gawk],
>>     [AC_PATH_PROGS_FEATURE_CHECK([MAKE], [make gmake],
>>     [AC_PATH_PROGS_FEATURE_CHECK([LIBTOOL], [libtool glibtool],
>>     [AC_PATH_PROGS_FEATURE_CHECK([LIBTOOLIZE], [libtoolize glibtoolize],
>>     [AC_PATH_PROGS_FEATURE_CHECK([automake], [automake],
>> --->8---
>>
>> Best regards
>>
>> Andreas Bießmann
>>
>>> On Tue, Mar 18, 2014 at 4:35 PM, Andreas Bießmann <andreas@biessmann.de> wrote:
>>>> Dear Ray Donelly,
>>>>
>>>> On 18.03.14 17:10, Ray Donnelly wrote:
>>>>> Why don't you use:
>>>>> configure --with-libtool=glibtool
>>>>
>>>> well, in fact this is the command I used to get it working in the first
>>>> place. I however think its worth to give something back to the community
>>>> and this is a trivial fix others should not stumble upon.
>>>>
>>>> Best regards
>>>>
>>>> Andreas Bießmann
>>>>
>>>>> On Tue, Mar 18, 2014 at 4:06 PM,  <andreas@biessmann.de> wrote:
>>>>>> # HG changeset patch
>>>>>> # User Andreas Bießmann <andreas@biessmann.de>
>>>>>> # Date 1395158470 -3600
>>>>>> #      Tue Mar 18 17:01:10 2014 +0100
>>>>>> # Node ID 2f1530b54afcb6a00e1d3ecc2595f588a3dd7315
>>>>>> # Parent  e11a8a2e225d3fa882c24e05ea097979ba8925eb
>>>>>> configure.ac: respect 'g' variants of libtool/libtoolize
>>>>>>
>>>>>> BSD OS'es (OS X for me) provide GNU tools with prefixed 'g'. To find correct
>>>>>> versions of libtool/libtoolize on those systems search also for
>>>>>> glibtool/glibtoolize.
>>>>>>
>>>>>> Signed-off-by: Andreas Bießmann <andreas@biessmann.de>
>>>>>>
>>>>>> diff --git a/configure.ac b/configure.ac
>>>>>> --- a/configure.ac
>>>>>> +++ b/configure.ac
>>>>>> @@ -242,7 +242,7 @@
>>>>>>                         [Specify the full PATH to GNU libtool >= 1.5.26]),
>>>>>>          [ac_cv_path_LIBTOOL=$withval])])
>>>>>>  AC_CACHE_CHECK([for GNU libtool >= 1.5.26], [ac_cv_path_LIBTOOL],
>>>>>> -    [AC_PATH_PROGS_FEATURE_CHECK([LIBTOOL], [libtool],
>>>>>> +    [AC_PATH_PROGS_FEATURE_CHECK([LIBTOOL], [libtool glibtool],
>>>>>>          [[LIBTOOL_ver=$($ac_path_LIBTOOL --version 2>&1 \
>>>>>>                          |$EGREP '\(GNU libtool.*\) (2[[:digit:]]*\.|1\.6[[:digit:]]*\.|1\.5\.[2-9][[:digit:]]+)')
>>>>>>            test -n "$LIBTOOL_ver" && ac_cv_path_LIBTOOL=$ac_path_LIBTOOL ac_path_LIBTOOL_found=:]],
>>>>>> @@ -258,7 +258,7 @@
>>>>>>                         [Specify the full PATH to GNU libtoolize >= 1.5.26]),
>>>>>>          [ac_cv_path_LIBTOOLIZE=$withval])])
>>>>>>  AC_CACHE_CHECK([for GNU libtoolize >= 1.5.26], [ac_cv_path_LIBTOOLIZE],
>>>>>> -    [AC_PATH_PROGS_FEATURE_CHECK([LIBTOOLIZE], [libtoolize],
>>>>>> +    [AC_PATH_PROGS_FEATURE_CHECK([LIBTOOLIZE], [libtoolize glibtoolize],
>>>>>>          [[LIBTOOLIZE_ver=$($ac_path_LIBTOOLIZE --version 2>&1 \
>>>>>>                          |$EGREP '\(GNU libtool.*\) (2[[:digit:]]*\.|1\.6[[:digit:]]*\.|1\.5\.[2-9][[:digit:]]+)')
>>>>>>            test -n "$LIBTOOLIZE_ver" && ac_cv_path_LIBTOOLIZE=$ac_path_LIBTOOLIZE ac_path_LIBTOOLIZE_found=:]],
>>>>
>>>
>>> --
>>> For unsubscribe information see http://sourceware.org/lists.html#faq
>>>
>>
>>
>> --
>> For unsubscribe information see http://sourceware.org/lists.html#faq
>>
>
> I was about to make a patch for this myself. So I'm in:

The only thing I forgot to mention is: do you plan to also do the same
for greadelf and any other 'g' prefixed tools? Maybe as a separate
patch?

> Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
>
> -Bryan

-Bryan

--
For unsubscribe information see http://sourceware.org/lists.html#faq

  reply	other threads:[~2014-03-18 20:56 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-03-18 16:02 andreas
2014-03-18 16:10 ` Ray Donnelly
2014-03-18 16:31   ` Andreas Bießmann
2014-03-18 16:39     ` Ray Donnelly
2014-03-18 16:51       ` Andreas Bießmann
2014-03-18 20:54         ` Bryan Hundven
2014-03-18 20:56           ` Bryan Hundven [this message]
2014-05-05 19:45 ` Yann E. MORIN

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CAJ+oik14akMHqLSq4sWwnTpeCkezOYZNt+YcD0U9CvV7-T3WsQ@mail.gmail.com \
    --to=bryanhundven@gmail.com \
    --cc=andreas@biessmann.de \
    --cc=crossgcc@sourceware.org \
    --cc=mingw.android@gmail.com \
    --cc=yann.morin.1998@free.fr \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).