autogen-users — For software developers that use AutoGen in their projects

[Autogen-users] disable local linking of libopts . . .

[Geof S. <Geo...@ut...>]

Hi Bruce (et. al),

We use AutoOpts for a project, and we bundle the library for linking.

My question is this:  how do we configure such that a local libopt library isn't used -- we only want to link to the bundled version we provide.  We're having problems with older versions popping up.

Thanks for your hard work!

Geof 
U of U, SoC, RA

Re: [Autogen-users] autogen "make check" Error

[Tom B. <tom...@gm...>]

On Wed, Sep 1, 2010 at 14:16, Bruce Korb <bru...@gm...> wrote:
> 5.6.3 is a wee bit long in the tooth. (It was old several years ago.)
> I'd recommend 5.10.2 or 5.11.1pre7:

Sorry, Bruce--I was asleep when I went to gnu.org and somehow missed
the later versions.

-Tom

Re: [Autogen-users] autogen "make check" Error

[Bruce K. <bru...@gm...>]

5.6.3 is a wee bit long in the tooth. (It was old several years ago.)
I'd recommend 5.10.2 or 5.11.1pre7:
  http://autogen.sourceforge.net/data/index.html

[Autogen-users] autogen "make check" Error

[Tom B. <tom...@gm...>]

I built autogen 5.6.3 successfully and then tried "make check" which
resulted in one failure:

PASS: argument.test
PASS: cond.test
FAIL: enums.test
PASS: equiv.test
PASS: errors.test
PASS: getopt.test
PASS: guile.test
PASS: handler.test
PASS: immediate.test
PASS: keyword.test
PASS: main.test
PASS: nls.test
PASS: rc.test
PASS: shell.test
PASS: stdopts.test
PASS: usage.test
PASS: vers.test
===========================================
1 of 17 tests failed
Please report to aut...@li...
===========================================

I am hereby reporting as ordered!

Regards,

-Tom

Thomas M. Browder, Jr.
Niceville, Florida
USA

[Autogen-users] last call on 5.11.1 pre-release

[Bruce K. <bru...@gm...>]

Hi,

I introduced some bugs with the new dependency stuff, so I have
to spin a new release.  So, I would really appreciate knowing
who-all has given it a spin.  My development environment just
cannot cover all the different platforms that everybody uses.
The "pre5" variation is available now:

   http://autogen.sourceforge.net/data/

Thank you very much!

Regards, Bruce

[Autogen-users] Support new platform "Andestech" Platform

[Thomas YS L. <nck...@gm...>]

Hi All


Patches are us following


---------- 轉寄的郵件 ----------
寄件者： <me...@an...>
日期： 2009年7月1日下午3:59
主旨： Andes application porting guide


 *Hi, all:* * *

Share you with the experiences in open source applications porting for Andes
platform.

* *

** Environment settings for Andes Toolchain*



# Andes toolchain setting

export ANDESIGHT_ROOT=/home/path/toolchains/nds32-elf-n1213-43u1h

export PATH=$ANDESIGHT_ROOT/bin:$PATH



# PREFIX folder

export PREFIX=/my/path/nds32



export CROSS_COMPILE="nds32le-linux-"

which ${CROSS_COMPILE}gcc &> /dev/null || export CROSS_COMPILE="nds32-elf-"



export HOST=${CROSS_COMPILE%-}

export BUILD="i686-linux"

export TARGET=${CROSS_COMPILE%-}



# Cross Toolchain

export BUILD_CC=gcc

export CC="${TARGET}-gcc"

export CXX="${TARGET}-g++"

export AR="${TARGET}-ar"

export AS="${TARGET}-as"

export RANLIB="${TARGET}-ranlib"

export LD="${TARGET}-ld"

export STRIP="${TARGET}-strip"



# Cross Compilation FLAGS (-EL is for little endian)

export NDS32_GCC_CFLAGS=-EL

#export CFLAGS="-EL -I$PREFIX/include"

#export CFLAGS="-EL -I$PREFIX/include -O2"

export CFLAGS="-EL -I$PREFIX/include -fomit-frame-pointer -funroll-loops
-O3"

# With Static Linking

#export LDFLAGS="-EL -static -L$PREFIX/lib"

With Shared Linking

export LDFLAGS="-EL -L$PREFIX/lib"

export ASFLAGS=-EL

export CXXFLAGS=-EL





note: The prefix of Andes toolchain is* nds32-elf* in AndeSight v1.3.x, but
we will modify the prefix to nds32le-linux since Andesight v1.3.3.



* Open Source Application Porting using Andes Toolchain



*1. Modify config.sub*
*1.1 Modify config.sub file* for individual package

This step helps the autoconf tool
<http://www.gnu.org/software/autoconf/>recognizing the Andes
architecture and machine type. You can follow the diff
file below to modify the corresponding field in your package’s config.sub
file.



*$ diff -u config.sub.orig config.sub*

*--- config.sub.orig     2009-03-03 14:45:26.000000000 +0800*

*+++ config.sub  2009-03-03 15:08:37.000000000 +0800*

*@@ -149,6 +149,7 @@*

* case $basic_machine in*

*        # Recognize the basic CPU types without company name.*

*        # Some are omitted here because they have special meanings below.*

*+       nds32 | nds32le \*

*        tahoe | i860 | m32r | m68k | m68000 | m88k | ns32k | arc | arm \*

*                | arme[lb] | pyramid | mn10200 | mn10300 \*

*                | tron | a29k | 580 | i960 | h8300 | hppa | hppa1.0 |
hppa1.1 \*

*@@ -171,6 +172,7 @@*

*                exit 1*

*                ;;*

*        # Recognize the basic CPU types with company name.*

*+       nds32-* | nds32le-* \*

*        vax-* | tahoe-* | i[3456]86-* | i860-* | m32r-* | m68k-* | m68000-*
\*

*              | m88k-* | sparc-* | ns32k-* | fx80-* | arc-* | arm-* |
c[123]* \*

*              | mips-* | pyramid-* | tron-* | a29k-* | romp-* | rs6000-* \*



or


*1.2 Modify config.sub file for automake*



Added NDS32 archicture from Andes Technology.

---

 config.sub |    2 ++

 1 file changed, 2 insertions(+)



Index: automake1.10-1.10.1/lib/config.sub

===================================================================

--- automake1.10-1.10.1.orig/lib/config.sub      2008-08-13
16:20:51.000000000 +0800

+++ automake1.10-1.10.1/lib/config.sub  2008-08-13 16:25:05.000000000 +0800

@@ -270,6 +270,7 @@

       | mn10200 | mn10300 \

       | mt \

       | msp430 \

+      *| nds32 | nds32le* \

       | nios | nios2 \

       | ns16k | ns32k \

       | or32 \

@@ -351,6 +352,7 @@

       | mmix-* \

       | mt-* \

       | msp430-* \

*+      | nds32-*|** nds32le-*** \*

       | nios-* | nios2-* \

       | none-* | np1-* | ns16k-* | ns32k-* \

       | orion-* \x



*2. configure and make*

./configure --host=$HOST --build=$BUILD --prefix=$PREFIX

make

#make install DESTDIR=${DEPLOY_DIR}

make install DESTDIR=${PREFIX}



Regards,

Mentor



-- 
Best Regards,

Thomas Yu - Sian , Liu

Re: [Autogen-users] Spurious whitespace in "dne" output

[Harlan S. <st...@nt...>]

Bruce,

I'm writing Kapila's eval now.

H

Re: [Autogen-users] Spurious whitespace in "dne" output

[Bruce K. <bk...@gn...>]

Hi Dennis,

On Thu, Jul 15, 2010 at 12:11 AM, Dennis Wassel
<dw...@ma...> wrote:
> Hi Bruce,
>
> I sent this to the list yesterday and got an auto-reply that I'd have to
> subscribe (which I did) and to give you a heads-up to fish this one out
> of your spam.
> So here you are!
>
> Cheers,
> Dennis
>
> -------- Original-Nachricht --------
> Betreff: [Autogen-users] Spurious whitespace in "dne" output
> Datum: Wed, 14 Jul 2010 17:23:21 +0200
> Von: Dennis Wassel <dw...@ma...>
> An: aut...@li...
>
> Hi list, hi Bruce,
>
> I am using the dne command like this
>
>  (dne "-d" " * " "/*")
[...]
> which is fine, except....
>
> Specifically, x86_64 Ubuntu 10.04's autogen (5.10) produces
>  "AutoGen-ed" (no whitespace)
> while i686 Debian5's autogen 5.9.5 and self-compiled 5.10.2 both insist
> on producing
>  "AutoGen-ed  " (2 whitespaces)
> so this always turns up in the diffs, which is nothing drastic, but
> still a PITA.

It isn't supposed to happen, as well as being a PITA.

> I hunted this down to something like a printf format string in
> agen5/expFormat.c, but I could not make much sense of it, particularly
> there are no two evil spaces hidden here.
>
> Any help?

This is the format:

static char const zDne[] = "%6$s"
"%1$sDO NOT EDIT THIS FILE   (%2$s)\n"
"%1$s\n"
"%1$sIt has been AutoGen-ed%3$s\n"
"%1$sFrom the definitions    %4$s\n"
"%1$sand the template file   %5$s";

I will certainly concede that how that format gets used is a bit obscure, but
the final result is basically:

static char const zDne1[] =
"%s -*- buffer-read-only: t -*- vi: set ro:\n"
"%s\n";
  asprintf(&str, zDne1, "/*", " * ")

  sprintf(zDne, " * ", out_fname, "", def_fname, tpl_fname, str);

Even more complicating is the fact that in order to compute printf arguments
on the fly, "snprintfv" was invented (an "argv" style printf library because
libffi was not around at the time).  So, it is possible that the snv_asprintfv
has gone south for this platform, but I'd surely hope not.  It has been stable
for some years now.

I hate to ask people to use GDB, but this is not something that is happening
for me, so I think it would be best, if you are willing.

Please remake the thing with CFLAGS='-g -O0' LDFLAGS=-static thus:

   cd $top_builddir/agen5
   rm -f *.o
   make autogen CFLAGS='-g -O0' LDFLAGS=-static
   dbgag=$PWD/autogen

cd into your build directory and:

   gdb --args $dbgag <<whatever>>

and set a break point in ag_scm_dne().  Step through until you get to
the line:
   pzRes = aprf(ENABLED_OPT(WRITABLE) .....
and examine the arguments on the Debian5 platform.

(gdb) p *pzPrefix
(gdb) p *pfp->pzOutName
(gdb) p (char *)zScribble
(gdb) p pz
(gdb) p pzTemplFileName
(gdb) p pzFirst

In particular, the ``p (char *)zScribble'' should print the empty string because
"noDate" should be set to 1.  Apparently, Debian5 is emitting two space
characters when it should not be.  If you see the spaces in this GDB session,
something is awry in agen5 code.  If not, then it is either in the snprintfv
code or it is further down in the bowels of Debian5.

Sorry for the bother.  I would surely appreciate understanding the issue!!

Regards, Bruce

[Autogen-users] Spurious whitespace in "dne" output

[Dennis W. <dw...@ma...>]

Hi list, hi Bruce,

I am using the dne command like this

  (dne "-d" " * " "/*")

to produce something like

  *
  * DO NOT EDIT THIS FILE   (foo.h)
  *
  * It has been AutoGen-ed
  * From the definitions    foo.def
  * and the template file   foo.tpl */

which is fine, except for some whitespace produced by some versions of
autogen, and not by others.

Specifically, x86_64 Ubuntu 10.04's autogen (5.10) produces
  "AutoGen-ed" (no whitespace)
while i686 Debian5's autogen 5.9.5 and self-compiled 5.10.2 both insist
on producing
  "AutoGen-ed  " (2 whitespaces)
so this always turns up in the diffs, which is nothing drastic, but
still a PITA.

I hunted this down to something like a printf format string in
agen5/expFormat.c, but I could not make much sense of it, particularly
there are no two evil spaces hidden here.

Any help?

Cheers,
Dennis

-- 
Zentrum für Technomathematik, AG Optimierung und Optimale Steuerung
 fon: 0421-218.63866   fax: 0421-218.9863866    web: www.worhp.de

[Autogen-users] An overdue enhancement

[Bruce K. <bru...@gm...>]

Hi,

I think someone on this list asked for this and I finally needed it, too.
I added make file dependency output using option names pretty similar to
those used by GCC:

  http://autogen.sourceforge.net/announce.html

Let me know if you have any problems:

  http://autogen.sourceforge.net/data/autogen-5.11pre1.tar.bz2

Cheers - Bruce

[Autogen-users] GNU AutoGen Version 5.10.1/AutoOpts Version 33.2

[Bruce K. <bk...@gn...>]

GNU AutoGen/AutoOpts is a two-part project that serves two separate
purposes.  The two parts are combined because they are inextricably
intertwined:

AutoGen is a tool designed to simplify the creation and maintenance of
programs that contain large amounts of repetitious text.  It is
especially valuable in programs that have several blocks of text that
must be kept synchronized.

AutoOpts is both an example of that and a project in its own right.
It is a very powerful configuration file, environment variable and
command line option management tool consisting of a set of AutoGen
templates and a run time library that nearly eliminates the hassle of
managing, parsing and documenting program options.

The self-referential example:
http://www.gnu.org/software/autogen/man1-autogen.html

New in 5.10.1 - March, 2010

Here are the 'NEWS' entries since the last release

* The "libguile" stuff has been removed.  I stopped using it and
  nobody else can have been using it.  It's gone.

* Fixed a seg fault that happens on some platforms after processing completes.

* Augmented the Finite State Machine templates to #include
  hand crafted callout code in addition to extracting from
  earlier revisions.

* autogen now uses unlocked io where feasible.

AutoGen home:    http://www.gnu.org/software/autogen/
primary ftp:     ftp://ftp.gnu.org/gnu/autogen/rel5.10.1/
.tar.gz:         ftp://ftp.gnu.org/gnu/autogen/rel5.10.1/autogen-5.10.1.tar.gz
library project: ftp://ftp.gnu.org/gnu/libopts/rel33.2/
bug reports:     autogen-users at the lists dot SourceForge net domain
bug archive:     http://sourceforge.net/mailarchive/forum.php?forum_id=7034
maintainer:      Bruce Korb - bkorb at the usual GNU domain

Re: [Autogen-users] Getting AutoGen working in Windows

[Aaron T. <syn...@gm...>]

On Sun, Feb 28, 2010 at 11:21 AM, Bruce Korb <bru...@gm...> wrote:
> Rowan Sylvester-Bradley wrote:
>> I'm trying to get AutoGen working under Windows Vista Business. I'm hampered
>> by the fact that I know almost nothing about Unix/Linux. I've downloaded and
>> installed Cygwin v1.7.1. It appears to run, although I don't know enough to
>> test it properly. Are there some test scripts or commands that I should try
>> to check that it's working properly?
>>
>> It also seems to have Guile installed - at least, I can type "guile" and get
>> a "guile>" prompt. I know nothing at all about Guile, so I don't know how to
>> test this either. Are there some scripts I should run or commands to try
>> that would check it out?
>>
>> To install AutoGen, what file(s) do I have to download, and what do I have
>> to do with them?
>
> Hi Rowan,
>
> You have to have libguile installed (you do) with the development headers
> as well (you may not).  I am forwarding this to the Minimalist GNU for Windows
> list.  They have something that works on Windows.  Unfortunately, Windows
> does not properly support fork() and guile uses some dynamic linking
> options that make its library problematic on Windows.  Both of these issues
> make Windows support challenging.  I don't use Windows myself.  So, hopefully,
> someone on the MinGW list can help point you in the right direction.  Sorry
> I could not be more helpful.
>
> Regards, Bruce

I don't know if this works with the latest version of AutoGen/Cygwin,
but I documented the necessary steps here:

http://tcpreplay.synfin.net/browser/trunk/docs/Win32Readme.txt

I probably last tested this a year ago.

-- 
Aaron Turner
http://synfin.net/
http://tcpreplay.synfin.net/ - Pcap editing and replay tools for Unix & Windows
Those who would give up essential Liberty, to purchase a little temporary
Safety, deserve neither Liberty nor Safety.
    -- Benjamin Franklin
"carpe diem quam minimum credula postero"

Re: [Autogen-users] Getting AutoGen working in Windows

[Bruce K. <bru...@gm...>]

Rowan Sylvester-Bradley wrote:
> I'm trying to get AutoGen working under Windows Vista Business. I'm hampered
> by the fact that I know almost nothing about Unix/Linux. I've downloaded and
> installed Cygwin v1.7.1. It appears to run, although I don't know enough to
> test it properly. Are there some test scripts or commands that I should try
> to check that it's working properly?
> 
> It also seems to have Guile installed - at least, I can type "guile" and get
> a "guile>" prompt. I know nothing at all about Guile, so I don't know how to
> test this either. Are there some scripts I should run or commands to try
> that would check it out?
> 
> To install AutoGen, what file(s) do I have to download, and what do I have
> to do with them?

Hi Rowan,

You have to have libguile installed (you do) with the development headers
as well (you may not).  I am forwarding this to the Minimalist GNU for Windows
list.  They have something that works on Windows.  Unfortunately, Windows
does not properly support fork() and guile uses some dynamic linking
options that make its library problematic on Windows.  Both of these issues
make Windows support challenging.  I don't use Windows myself.  So, hopefully,
someone on the MinGW list can help point you in the right direction.  Sorry
I could not be more helpful.

Regards, Bruce

[Autogen-users] Getting AutoGen working in Windows

[Rowan Sylvester-B. <ro...@sy...>]

I'm trying to get AutoGen working under Windows Vista Business. I'm hampered
by the fact that I know almost nothing about Unix/Linux. I've downloaded and
installed Cygwin v1.7.1. It appears to run, although I don't know enough to
test it properly. Are there some test scripts or commands that I should try
to check that it's working properly?

It also seems to have Guile installed - at least, I can type "guile" and get
a "guile>" prompt. I know nothing at all about Guile, so I don't know how to
test this either. Are there some scripts I should run or commands to try
that would check it out?

To install AutoGen, what file(s) do I have to download, and what do I have
to do with them?

Thanks for your help - Rowan

[Autogen-users] Autogen for Windows

[Rowan Sylvester-B. <ro...@sy...>]

Does anyone have a Windows executable for AutoGen, or blow-by-blow 
instructions for getting it working in Windows?

Thanks - Rowan 

Re: [Autogen-users] missing

[Eric B. <eb...@by...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

According to Bruce Korb on 10/18/2009 12:21 PM:
> Eric Blake wrote:
>> But on Windows 64 bit platforms, that is exactly the case (64-bit
>> pointers, 32-bit long).
> 
> Thanks, Eric.  More compat.h magic:
> 
>> #ifndef HAVE_INTPTR_T
>> # if SIZEOF_CHARP == SIZEOF_LONG
>>     typedef signed long         intptr_t;

On any platform that lacks intptr_t, this should be right - the only
platforms stupid enough to provide 32-bit long but 64-bit char* (aka
windows 64) should at least be sane enough to provide intptr_t.

>> # else
>>     typedef signed int          intptr_t;

This arm of the #if is pointless.  On windows 64, int and long are the
same size.  On almost every other platform, char* and long are the same
size, and either int and long are the same size (32 bit platforms) or int
is smaller than pointers (64 bit platforms).

- --
Don't work too hard, make some time for fun as well!

Eric Blake             eb...@by...
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Cygwin)
Comment: Public key at home.comcast.net/~ericblake/eblake.gpg
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkrbhBMACgkQ84KuGfSFAYCjTACgrr02LIrudj/HD0VRo1cjTPpe
nSIAniPrEV+hE1+8DhvGFL4WkR2IUUTn
=UPZg
-----END PGP SIGNATURE-----

Re: [Autogen-users] missing

[Bruce K. <bk...@gn...>]

Hi Eric,

On Sun, Oct 18, 2009 at 2:09 PM, Eric Blake <eb...@by...> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> According to Bruce Korb on 10/18/2009 12:21 PM:
>> Eric Blake wrote:
>>> But on Windows 64 bit platforms, that is exactly the case (64-bit
>>> pointers, 32-bit long).
>>
>> Thanks, Eric.  More compat.h magic:
>>
>>> #ifndef HAVE_INTPTR_T
>>> # if SIZEOF_CHARP == SIZEOF_LONG
>>>     typedef signed long         intptr_t;
>
> On any platform that lacks intptr_t, this should be right - the only
> platforms stupid enough to provide 32-bit long but 64-bit char* (aka
> windows 64) should at least be sane enough to provide intptr_t.

What started this all off is that there was some platform out there
that did not define intptr_t and had a pointer size that was not that
of long.  Not having access to such a platform (and being uncertain
that I'd even _want_ to ;), I cooked up this stuff so NTP would build
on whatever it was that was that broken.  My hope was to try to
shoot the problem dead so I never see it again.  :)

Thank you!  Cheers - Bruce

Re: [Autogen-users] missing

[Bruce K. <bk...@gn...>]

Eric Blake wrote:
> 
> But on Windows 64 bit platforms, that is exactly the case (64-bit
> pointers, 32-bit long).

Thanks, Eric.  More compat.h magic:

> #ifndef HAVE_INTPTR_T
> # if SIZEOF_CHARP == SIZEOF_LONG
>     typedef signed long         intptr_t;
> # else
>     typedef signed int          intptr_t;
> # endif
> #endif
> 
> #ifndef HAVE_UINTPTR_T
> # if SIZEOF_CHARP == SIZEOF_LONG
>     typedef unsigned long       intptr_t;
> # else
>     typedef unsigned int        intptr_t;
> # endif
> #endif

This is in 5.9.10pre12
see: http://autogen.sourceforge.net/data/

Regards, Bruce

Re: [Autogen-users] missing

[Bruce K. <bk...@gn...>]

On Tue, Oct 13, 2009 at 5:52 PM, Eric Blake <eb...@by...> wrote:
>> then the backup is "unsigned long".  If a pointer is
>> not matching a long in size, there is a terrible lot of
>> code that will break.
>
> But on Windows 64 bit platforms, that is exactly the case (64-bit
> pointers, 32-bit long).

Hmmm.  Well, I haven't done any development on windows
since Windows 3.1.  I read that misstatement and believed it
because it was consistent with the memory models I've seen
on UNIXy platforms.  Windows does seem to go out of its way
to be peculiar.  That will just make the tweaking of the
installed options.h header a tiny bit more complicated.  Oh, well.
Thanks.

Re: [Autogen-users] missing

[Eric B. <eb...@by...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

According to Bruce Korb on 10/13/2009 3:15 PM:
> If it isn't in stdint.h (it should be, if memory serves),
> then the backup is "unsigned long".  If a pointer is
> not matching a long in size, there is a terrible lot of
> code that will break.

But on Windows 64 bit platforms, that is exactly the case (64-bit
pointers, 32-bit long).

- --
Don't work too hard, make some time for fun as well!

Eric Blake             eb...@by...
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Cygwin)
Comment: Public key at home.comcast.net/~ericblake/eblake.gpg
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkrVIN4ACgkQ84KuGfSFAYCOvQCdFacdhewnQw+MuVo+NGZZH/Xl
wiAAoMVlgfKy2VUX85YSdY+taRz3aZOn
=cyMy
-----END PGP SIGNATURE-----

Re: [Autogen-users] missing

[Bruce K. <bk...@gn...>]

If it isn't in stdint.h (it should be, if memory serves),
then the backup is "unsigned long".  If a pointer is
not matching a long in size, there is a terrible lot of
code that will break.  So, in compat/compat.h:

#ifndef HAVE_UINTPTR_T
typedef unsigned long uintptr_t;
#endif

I guess that needs to get configured into the installed header.
What a pain:

../libopts/autoopts/options.h:348: error: expected
specifier-qualifier-list before ‘uintptr_t’

I have a script for installing an edited version of options.h.
It needs to edit in the above typedef if HAVE_UINTPTR_T
is not defined in config.h.  *sigh*.  Thank you.

[Autogen-users] missing

[Geof S. <Geo...@ut...>]

/*   -*- buffer-read-only: t -*- vi: set ro:
 *  
 *  DO NOT EDIT THIS FILE   (sched-opt.h)
 *  
 *  It has been AutoGen-ed  Thursday October  8, 2009 at 11:18:06 AM MDT
 *  From the definitions    sched-opt.def
 *  and the template file   options
 *
 * Generated from AutoOpts 32:0:7 templates.
 */

/*
 *  This file was produced by an AutoOpts template.  AutoOpts is a
 *  copyrighted work.  This header file is not encumbered by AutoOpts
 *  licensing, but is provided under the licensing terms chosen by the
 *  isp author or copyright holder.  AutoOpts is licensed under
 *  the terms of the LGPL.  The redistributable library (``libopts'') is
 *  licensed under the terms of either the LGPL or, at the users discretion,
 *  the BSD license.  See the AutoOpts and/or libopts sources for details.
 *
 * This source file is copyrighted and licensed under the following terms:
 *
 * isp copyright 1-27-2009 University of Utah - all rights reserved
 *
 * Copyright (C) 2009 -- please see LICENSE for details.
 */
/*
 *  This file contains the programmatic interface to the Automated
 *  Options generated for the isp program.
 *  These macros are documented in the AutoGen info file in the
 *  "AutoOpts" chapter.  Please refer to that doc for usage help.
 */
#ifndef AUTOOPTS_SCHED_OPT_H_GUARD
#define AUTOOPTS_SCHED_OPT_H_GUARD 1
#include <autoopts/options.h>

/*
 *  Ensure that the library used for compiling this generated header is at
 *  least as new as the version current when the header template was released
 *  (not counting patch version increments).  Also ensure that the oldest
 *  tolerable version is at least as old as what was current when the header
 *  template was released.
 */
#define AO_TEMPLATE_VERSION 131072
#if (AO_TEMPLATE_VERSION < OPTIONS_MINIMUM_VERSION) \
 || (AO_TEMPLATE_VERSION > OPTIONS_STRUCT_VERSION)
# error option template version mismatches autoopts/options.h header
  Choke Me.
#endif

/*
 *  Enumeration of each option:
 */
typedef enum {
        INDEX_OPT_NUMPROCS         =  0,
        INDEX_OPT_PORT             =  1,
        INDEX_OPT_HOST             =  2,
        INDEX_OPT_US               =  3,
        INDEX_OPT_BLOCK            =  4,
        INDEX_OPT_NOBLOCK          =  5,
        INDEX_OPT_LOGFILE          =  6,
        INDEX_OPT_MPICALLS         =  7,
        INDEX_OPT_VERBOSE          =  8,
        INDEX_OPT_QUIET            =  9,
        INDEX_OPT_RPT_PROGRESS     = 10,
        INDEX_OPT_DISTRIBUTED      = 11,
        INDEX_OPT_FIBOPT           = 12,
        INDEX_OPT_DISABLE_OMP      = 13,
        INDEX_OPT_EXP_MODE         = 14,
        INDEX_OPT_ENV              = 15,
        INDEX_OPT_EXP_SOME         = 16,
        INDEX_OPT_STOP_AT_DEADLOCK = 17,
        INDEX_OPT_VERSION          = 18,
        INDEX_OPT_HELP             = 19,
        INDEX_OPT_MORE_HELP        = 20
} teOptIndex;

#define OPTION_CT    21
#define ISP_VERSION       "0.1.1"
#define ISP_FULL_VERSION  "isp - isp: In-Situ Partial Order - Ver. 0.1.1"

/*
 *  Interface defines for all options.  Replace "n" with the UPPER_CASED
 *  option name (as in the teOptIndex enumeration above).
 *  e.g. HAVE_OPT( NUMPROCS )
 */
#define         DESC(n) (ispOptions.pOptDesc[INDEX_OPT_## n])
#define     HAVE_OPT(n) (! UNUSED_OPT(& DESC(n)))
#define      OPT_ARG(n) (DESC(n).optArg.argString)
#define    STATE_OPT(n) (DESC(n).fOptState & OPTST_SET_MASK)
#define    COUNT_OPT(n) (DESC(n).optOccCt)
#define    ISSEL_OPT(n) (SELECTED_OPT(&DESC(n)))
#define ISUNUSED_OPT(n) (UNUSED_OPT(& DESC(n)))
#define  ENABLED_OPT(n) (! DISABLED_OPT(& DESC(n)))
#define  STACKCT_OPT(n) (((tArgList*)(DESC(n).optCookie))->useCt)
#define STACKLST_OPT(n) (((tArgList*)(DESC(n).optCookie))->apzArgs)
#define    CLEAR_OPT(n) STMTS( \
                DESC(n).fOptState &= OPTST_PERSISTENT_MASK;   \
                if ( (DESC(n).fOptState & OPTST_INITENABLED) == 0) \
                    DESC(n).fOptState |= OPTST_DISABLED; \
                DESC(n).optCookie = NULL )

/* * * * * *
 *
 *  Interface defines for specific options.
 */
#define VALUE_OPT_NUMPROCS       'n'
#define OPT_VALUE_NUMPROCS       (DESC(NUMPROCS).optArg.argInt)
#define VALUE_OPT_PORT           'p'
#define OPT_VALUE_PORT           (DESC(PORT).optArg.argInt)
#define VALUE_OPT_HOST           'h'
#ifndef WIN_32
#define VALUE_OPT_US             'x'
#endif /* WIN_32 */
#define VALUE_OPT_BLOCK          'b'
#define VALUE_OPT_NOBLOCK        'g'
#define VALUE_OPT_LOGFILE        'l'
#define VALUE_OPT_MPICALLS       'm'
#define VALUE_OPT_VERBOSE        'O'
#define VALUE_OPT_QUIET          'q'
#define VALUE_OPT_RPT_PROGRESS   'r'
#define OPT_VALUE_RPT_PROGRESS   (DESC(RPT_PROGRESS).optArg.argInt)
#define VALUE_OPT_DISTRIBUTED    'd'
#ifdef FIB
#define VALUE_OPT_FIBOPT         'f'
#endif /* FIB */
#ifdef USE_OPENMP
#define VALUE_OPT_DISABLE_OMP    'o'
#endif /* USE_OPENMP */
typedef enum {
    EXP_MODE_ALL,         EXP_MODE_RANDOM,      EXP_MODE_LEFT_MOST
} te_Exp_Mode;
#define VALUE_OPT_EXP_MODE       'e'
#define OPT_EXP_MODE_VAL2STR(_v) optionKeywordName( &DESC(EXP_MODE), (_v))
#define OPT_VALUE_EXP_MODE       (DESC(EXP_MODE).optArg.argEnum)
#define VALUE_OPT_ENV            's'
#define VALUE_OPT_EXP_SOME       'y'
#define OPT_VALUE_EXP_SOME       (DESC(EXP_SOME).optArg.argInt)
#define VALUE_OPT_STOP_AT_DEADLOCK 'z'
#define OPT_VALUE_STOP_AT_DEADLOCK (DESC(STOP_AT_DEADLOCK).optArg.argBool)
#define VALUE_OPT_HELP          '?'
#define VALUE_OPT_MORE_HELP     '!'
#define VALUE_OPT_VERSION       'v'
/*
 *  Interface defines not associated with particular options
 */
#define ERRSKIP_OPTERR  STMTS( ispOptions.fOptSet &= ~OPTPROC_ERRSTOP )
#define ERRSTOP_OPTERR  STMTS( ispOptions.fOptSet |= OPTPROC_ERRSTOP )
#define RESTART_OPT(n)  STMTS( \
                ispOptions.curOptIdx = (n); \
                ispOptions.pzCurOpt  = NULL )
#define START_OPT       RESTART_OPT(1)
#define USAGE(c)        (*ispOptions.pUsageProc)( &ispOptions, c )
/* extracted from /usr/share/autogen/opthead.tpl near line 415 */

/* * * * * *
 *
 *  Declare the isp option descriptor.
 */
#ifdef  __cplusplus
extern "C" {
#endif

extern tOptions   ispOptions;

#if defined(ENABLE_NLS)
# ifndef _
#   include <stdio.h>
    static inline char* aoGetsText( char const* pz ) {
        if (pz == NULL) return NULL;
        return (char*)gettext( pz );
    }
#   define _(s)  aoGetsText(s)
# endif /* _() */

# define OPT_NO_XLAT_CFG_NAMES  STMTS(ispOptions.fOptSet |= \
                                    OPTPROC_NXLAT_OPT_CFG;)
# define OPT_NO_XLAT_OPT_NAMES  STMTS(ispOptions.fOptSet |= \
                                    OPTPROC_NXLAT_OPT|OPTPROC_NXLAT_OPT_CFG;)

# define OPT_XLAT_CFG_NAMES     STMTS(ispOptions.fOptSet &= \
                                  ~(OPTPROC_NXLAT_OPT|OPTPROC_NXLAT_OPT_CFG);)
# define OPT_XLAT_OPT_NAMES     STMTS(ispOptions.fOptSet &= \
                                  ~OPTPROC_NXLAT_OPT;)

#else   /* ENABLE_NLS */
# define OPT_NO_XLAT_CFG_NAMES
# define OPT_NO_XLAT_OPT_NAMES

# define OPT_XLAT_CFG_NAMES
# define OPT_XLAT_OPT_NAMES

# ifndef _
#   define _(_s)  _s
# endif
#endif  /* ENABLE_NLS */

#ifdef  __cplusplus
}
#endif
#endif /* AUTOOPTS_SCHED_OPT_H_GUARD */
/* sched-opt.h ends here */

Re: [Autogen-users] autoopt and Windows . . .

[Earnie B. <ea...@us...>]

Quoting Geof Sawaya <Geo...@ut...>:

> Hi Aaron,
>
> Thanks for your pointers on this, and congrats on your tcpreplay project.
>
> I'm attempting to do this on the MinGW path so that we can  
> distribute without Cygwin dependencies . . .
>
> I'll let you know how it goes!
>

You won't get by without MSYS at least.  Especially for the scripts.   
And, IIRC, it must be created in the MSYS build environment, check the  
MinGW wiki for details on that and ask on  
min...@li....

--
Earnie

Re: [Autogen-users] autoopts 'option template version mismatch'

[Bruce K. <bk...@gn...>]

It means you are finding an autoopts/options.h header
that does not match the template you are using.
Use "-H" on the CC line to figure out where it is coming
from and make sure you have a -I that points to the
new one ahead of wherever you are getting the old one.

On Fri, Oct 9, 2009 at 11:54 AM, Geof Sawaya <Geo...@ut...> wrote:
> Hello,
>
> I’m experiencing this template version mismatch.  (the below #error
> condition is being hit)
>
> Does anyone have a quick idea what causes this?  Thanks in advance for any

[Autogen-users] autoopts 'option template version mismatch'

[Geof S. <Geo...@ut...>]

Hello,

I'm experiencing this template version mismatch.  (the below #error condition is being hit)

Does anyone have a quick idea what causes this?  Thanks in advance for any help.



****from my generated *-opt.h file*************

#define AO_TEMPLATE_VERSION 131072
#if (AO_TEMPLATE_VERSION < OPTIONS_MINIMUM_VERSION) \
 || (AO_TEMPLATE_VERSION > OPTIONS_STRUCT_VERSION)
# error option template version mismatches autoopts/options.h header
  Choke Me.
#endif