From dclarke at blastwave.org  Mon Mar 24 17:47:42 2008
From: dclarke at blastwave.org (Dennis Clarke)
Date: Mon, 24 Mar 2008 12:47:42 -0400 (EDT)
Subject: [csw-dev] test the dev list !
Message-ID: <2283.72.39.133.97.1206377262.squirrel@mail.blastwave.org>





        THIS  PAGE  INTENTIONALLY  LEFT  BLANK
           yes .. a paradox if it were true


From jeff at cjsa.com  Mon Mar 24 18:13:02 2008
From: jeff at cjsa.com (Jeffery Small)
Date: Mon, 24 Mar 2008 17:13:02 GMT
Subject: [csw-dev] What's this list for?
Message-ID: <Jy8vtq.Iqx@cjsa.com>

Sorry, I must have missed some discussion along the way.  What is the dev list
supposed to be for as opposed to the maint list?  Thanks.

Regards,
-- 
Jeff

Jeffery Small  <jeff at blastwave.org>
In real life:  <jeff at cjsa.com>



From dam at blastwave.org  Mon Mar 24 21:48:48 2008
From: dam at blastwave.org (Dagobert Michelsen)
Date: Mon, 24 Mar 2008 21:48:48 +0100
Subject: [csw-dev] Multi-ISA GAR
In-Reply-To: <47E805E3.1060705@pleiades.fr.eu.org>
References: <20080319100038.C3150@bolthole.com>
	<33F3A8F3-6FA2-4E2E-91D2-6909A9902BCC@blastwave.org>
	<47E805E3.1060705@pleiades.fr.eu.org>
Message-ID: <F5645699-8B12-4DD0-BA05-FD383BE95C47@blastwave.org>

Hi Yann,

Am 24.03.2008 um 20:49 schrieb Yann Rouillard:
> I began to convert openssl to the new multi-ISA system, I have two
> comments so far:
>
> 	- the default isa install target is fine for standard programs  
> using automatic-multiarch-compile
> the classic configure script but will not work for everything (and
>
> indeed not for openssl).
> 	I added a new INSTALL_EXTRA_ISA_SCRIPTS to allow to redefine the  
> target
> used: see http://svn.blastwave.org/trac/changeset/1994

This is a good solution for openssl, but I don't believe
that there are any more packages built in the same way
(correct me if I'm wrong on this). I think it would be better
to solve this with

ifneq ($(ISA),$(ISA_DEFAULT))
INSTALL_SCRIPTS = $(INSTALL_EXTRA_ISA_SCRIPTS)
endif

inside the openssl-Makefile and not make a general rule out
of it. A more general solution would be

   INSTALL_SCRIPTS_$(ISA) ?= $(WORKSRC)/Makefile
   INSTALL_SCRIPTS ?= $(INSTALL_SCRIPTS_ISA)

with overwriting the specific install scripts on demand.
The thing that is disturbing me is the special handling of the default
ISA. In fact there are two default-ISAs: one for 32 bit and one for  
64 bit.
This makes defining the extra-ISAS a bit ugly IMHO.

> 	- I would like to be able to not use the automatic prototype  
> generation
> but I didn't see a way to disable your new automatic prototype
> generation. Could you add this option ?

This should work out-of-the-box. If the prototype is defined
in DISTFILES it shouldn't get overridden, so existing Makefiles
shouldn't break. Generally dynamic prototypes don't need adjustment
on version updates, so if there is anything missing which you need to
use them let me know what is missing and we may find a general solution.

> 	- I still have a problem building openssl, the first time I try to  
> build the package target after a "gmake clean", I have the  
> following error:
> 	gmake: *** No rule to make target `package-CSWossl', needed by  
> `package'.  Stop.
> 	If retry "gmake package" after this error, it works.
>
> 	You'll find the debug log of the first gmake call in my home:  
> ~yann/gmake_package.debug.log
> 	You'll find my automatic-multiarch-compile build tree on ra: ~yann/ 
> csw2/automatic-multiarch-compile/

This seems to be the same issue: dynamic prototypes. I'll make
a test environment with your openssl-Makefile and try to reproduce
this.


Best regards

   -- Dago


From dam at blastwave.org  Mon Mar 24 21:52:29 2008
From: dam at blastwave.org (Dagobert Michelsen)
Date: Mon, 24 Mar 2008 21:52:29 +0100
Subject: [csw-dev] What's this list for?
In-Reply-To: <Jy8vtq.Iqx@cjsa.com>
References: <Jy8vtq.Iqx@cjsa.com>
Message-ID: <5FC3DC1E-D19D-44F0-9E7E-1257BABC9A89@blastwave.org>

Hi Jeff,

Am 24.03.2008 um 17:13 schrieb Jeffery Small:
> Sorry, I must have missed some discussion along the way.  What is  
> the dev list
> supposed to be for as opposed to the maint list?  Thanks.

This list is for everything concerning package builds,
from compiler-flags to GAR, from pkg-get to Trac.
The only thing which should remain on the maintainer
list are discussions about people joining and leaving
and other things which may be better discussed in a
closed group.

The idea is that this dev-list is open for everybody
with an open mailing list archive while the maintainer
list stays closed to active maintainers.


Best regards

   -- Dago


From dam at blastwave.org  Mon Mar 24 22:20:05 2008
From: dam at blastwave.org (Dagobert Michelsen)
Date: Mon, 24 Mar 2008 22:20:05 +0100
Subject: [csw-dev] [libart 0002824]: RPATH contains
	/opt/csw/lib/sparcv9/$ISALIST
In-Reply-To: <20080324.20525100.1688352647@gyor.oxdrove.co.uk>
References: <200803211925.m2LJP8YB003508@callisto.blastwave.org>
	<20080321.19493100.2182619288@gyor.oxdrove.co.uk>
	<6C4B6E12-C46B-41F9-B1E3-7BFBFD8D8A9C@blastwave.org>
	<20080321.20314500.1513284418@gyor.oxdrove.co.uk>
	<16E0CBC5-1471-4C5C-818C-E27561981246@blastwave.org>
	<20080321.20470900.4124343341@gyor.oxdrove.co.uk>
	<836BD06B-8EF4-4148-87D8-51C6E10F070C@blastwave.org>
	<20080321.21201000.1964856037@gyor.oxdrove.co.uk>
	<C338E757-CD81-4093-AE1C-44FDA6BAC599@blastwave.org>
	<20080322.11171500.2414617030@gyor.oxdrove.co.uk>
	<75F0377E-79EE-47D6-983A-B2776010056E@blastwave.org>
	<20080324.20525100.1688352647@gyor.oxdrove.co.uk>
Message-ID: <7A398180-54F8-4E7F-B524-BDE40EA56EF4@blastwave.org>

Hi James.

I moved this to the new dev-list, as it may be worth
conserving the discussion :-) That's why I am keeping
most of the discussion fully quoted.

Am 24.03.2008 um 20:52 schrieb James Lee:
> On 24/03/08, 19:48:09, Dagobert Michelsen <dam at blastwave.org> wrote
>>>   export OPT="-fast -xnolibmopt -xspace -xtarget=ultra2 -xarch=v8
>>> -xstrconst"
>>>   make -e clean all
>>>   mv cswstage cswstage.generic
>>>
>>>   export OPT="-fast -xtarget=ultra3i -xarch=v8plusb -xstrconst"
>>>   export MLIB="-lm"
>>>   make -e clean all
>>>   mv cswstage cswstage.sparcv9+vis2
>
>> Ok, so you have different optimizer flags for each ISA, right?
>> And you have different exported variables per ISA.
>
> Yes.  The flags are not just about matching the instruction set.
> The flags are to suit the needs of the binaries.  Eg, a rarely used
> utility or one used for < 100ms is best optimised for size and
> certainly doesn't need arch variants.  The binaries that run for
> hours get everything I can give them.  It's not even true that the
> expected flags work best.
>
> I'm even putting v8 files in the v9 isaexec directory because
> Solaris 8 lacks the v8plusb isaexec entry (works with bins, not libs).

Looks quite like a hack... Any idea why the libs don't work?
I always thought the linker inspected the libs and decided
if the memory model is suitable instead of looking at the
directory name.

>> Why do you
>> use libm only for certain ISAs?
>
> Size optimisation.  It's really the inline libs I'm controlling
> here.

I see. So it would make sense to you to control the exported
variables per ISA?

>>> make actual calls a makefile that calls a special rmakefile, I  
>>> digress
>>> but somewhere in GAR it said make had to be gmake - whoa!!!
>
>> I'm sorry, I can't follow you here. Is gmake a problem?
>
> Yes when a makefile is written for make.  Now I could rewrite the
> makefile or I could use old make - easy choice.

If it is needed we can provide rules for ccsmake and xpg4make
which can easily selected if that helps you.

> Are yes, do we
> have imake capabilities?

Yes, those were inherited:

> # configure a package that uses imake
> # FIXME: untested and likely not the right way to handle the
> # argumentsconfigure-%/Imakefile:
>         @echo " ==> Running xmkmf in $*"
>         @cd $* && $(CONFIGURE_ENV) xmkmf $(CONFIGURE_ARGS)
>         @$(MAKECOOKIE)



> Maintainers with packages that just
> "configure && gmake install" are very lucky.

You can easily change behaviour to suit your needs with
   CONFIGURE_SCRIPTS = custom
   BUILD_SCRIPTS = custom

configure-custom: ...

build-custom: ...

>>> ARCHES="sparcv8 sparcv8plus+vis sparcv9+vis2"
>>> PROGS="rpict rvu rtrace ranimove ximage pfilt"
>>>
>>> for arch in $ARCHES
>>> do
>>>   mkdir -p `pwd`/cswstage/opt/csw/bin/$arch
>>>   for prog in $PROGS
>>>   do
>>>     # repeat in arch loop for simplicity...
>>>     rm -f cswstage/opt/csw/bin/$prog
>>>     cp cswstage.$arch/opt/csw/bin/$prog cswstage/opt/csw/bin/$arch/
>>> $prog
>>>   done
>>> done
>
>> Does PROGS contain bin/* or is it a subset?
>
> Just those 6, of about 125.  I definitely don't want 4 arch examples
> all.

Okay.

>> Additionally, you configure the binaries all to standard
>> locations (like bin/) and then move them to the ISA
>> subdirs. In the current multi-ISA GAR the individual
>> ISA-builds are configured to already use the ISA-subdirs
>> (like bindir=bin/sparcv9) which looks cleaner to me.
>
> That's where the current makefile puts them, it's not worth fighting,
> just go with the flow and move after.  There is no point it changing
> the build (and passing back) because only a packager wants the options
> and isaexec, anyone building for themselves knows their target.

Yes, but currently GAR resets bindir, libdir, etc. to the
ISA subdirectories automatically. This introduces some complexity
which I thought was necessary, but I am doubting now if this
could be reverted to the style you use. This would make the
GAR design quite a bit simpler.

> I know it takes the machine longer to build 5 variants and then
> only use a small number of each but the machine's time is cheap,
> mine is short (even if it's cheap ;-) ).
>
>>> ie, "manually", in this case it is almost regular but still a  
>>> country
>>> mile from wanting each example of each arch.  I really can't see an
>>> easier and clearer way of expressing these actions.
>
>> The steps are fairly straight-forward, and so a decent
>> abstraction could be added to GAR. I'll prepare something
>> for further discussion.
>
> Just keep in mind the need for flexibility.   I'm not against
> automation and build system in principle - the first thing I do with
> any project is waste 50% of the time working out how to do the job
> twice as quickly.  I am against systems that don't help.

I hereby allow you to kick me as long as you think
GAR doesn't help you ;-)

>>> For special optimisation, simplistically either speed or
>>> space, I want to be able use whichever binary benefits from each
>>> set of
>>> flags.
>
>> I see. Thanks for the input. Are you currently doing
>> these steps manually or do you have a script?
>
> I just record these steps in a script file, then run it.  It's all
> logged and repeatable and kept for the required 3 GPL years.
>
>
>
>> I would
>> like to convert this to GAR as a reference for my
>> "5 non-trivial packages" as it uses an extreme diversity
>> in compiler flags and ISA. As you have done quite some
>> work with compiler optiomization flags it would be great
>> if you could review the optimizer flag section in the
>> multi-ISA gar.conf.mk:
>>    <http://svn.blastwave.org/trac/browser/csw/branches/dam/automatic-
>> multiarch-compile/gar.conf.mk#L261>
>
>
> How about povray where I do profiling runs and feedback that back
> for recompilation?
>
> build(){
>
> for makefile in $(find . -name Makefile)
> do
>     gsed -i -e '/^CXXFLAGS.*/s//CXXFLAGS='"$FLAGS"' - 
> xprofile=collect/'
> $makefile
> done
>
>
> rm -rf povray.profile
> gmake clean
> gmake -j $(psrinfo | wc -l)
>
> for i in $(find scenes -name \*.pov)
> do
>     echo $i
>     echo ${i%/*}
>     unix/povray -D +A +L ./include/ +L ${i%/*}/ $i
> done
>
> for makefile in $(find . -name Makefile)
> do
>    gsed -i -e 's~^CXXFLAGS.*~CXXFLAGS='"$FLAGS"'
> -xprofile=use:'$(pwd)'\/povray~' $makefile
> done
>
> gmake clean
> gmake -j $(psrinfo | wc -l)
>
> mkdir -p cswstage/opt/csw/bin/${ARCH}
> cp unix/povray cswstage/opt/csw/bin/${ARCH}
>
> }
>
> FLAGS='-fast -xnolibmopt -xspace -xtarget=ss20 -xarch=v8'
> ARCH=sparcv8
> build
>
> FLAGS='-fast -xtarget=ultra2'
> ARCH=sparcv8plus+vis
> build
>
> if isalist | grep vis2 ; then
> FLAGS='-fast -xtarget=ultra3i -xarch=v8plusb'
> ARCH=sparcv9+vis2
> build
> fi

Fine. I'll give it a try when libiconv is finished and GAR
gets refined.


Thanks!

   -- Dago


From ihsan at blastwave.org  Mon Mar 24 22:36:50 2008
From: ihsan at blastwave.org (Ihsan Dogan)
Date: Mon, 24 Mar 2008 22:36:50 +0100
Subject: [csw-dev] What's this list for?
In-Reply-To: <Jy8vtq.Iqx@cjsa.com>
References: <Jy8vtq.Iqx@cjsa.com>
Message-ID: <47E81EF2.8040309@blastwave.org>

Am 24.3.2008 18:13 Uhr, Jeffery Small schrieb:
> Sorry, I must have missed some discussion along the way.  What is the dev list
> supposed to be for as opposed to the maint list?  Thanks.
>   
We were discussing that a few weeks ago on the maintainers list. Because 
I was abroad for two weeks, I didn't had any time to take care of this.

https://lists.blastwave.org/mailman/private/maintainers/2008-March/021245.html



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/



From yann at pleiades.fr.eu.org  Mon Mar 24 22:53:40 2008
From: yann at pleiades.fr.eu.org (Yann Rouillard)
Date: Mon, 24 Mar 2008 22:53:40 +0100
Subject: [csw-dev] Multi-ISA GAR
In-Reply-To: <F5645699-8B12-4DD0-BA05-FD383BE95C47@blastwave.org>
References: <20080319100038.C3150@bolthole.com>
	<33F3A8F3-6FA2-4E2E-91D2-6909A9902BCC@blastwave.org>
	<47E805E3.1060705@pleiades.fr.eu.org>
	<F5645699-8B12-4DD0-BA05-FD383BE95C47@blastwave.org>
Message-ID: <47E822E4.6080608@pleiades.fr.eu.org>

Dagobert Michelsen a ?crit :
> Hi Yann,
> 
> Am 24.03.2008 um 20:49 schrieb Yann Rouillard:
>> I began to convert openssl to the new multi-ISA system, I have two
>> comments so far:
>>
>>     - the default isa install target is fine for standard programs 
>> using automatic-multiarch-compile
>> the classic configure script but will not work for everything (and
>>
>> indeed not for openssl).
>>     I added a new INSTALL_EXTRA_ISA_SCRIPTS to allow to redefine the 
>> target
>> used: see http://svn.blastwave.org/trac/changeset/1994
> 
> This is a good solution for openssl, but I don't believe
> that there are any more packages built in the same way
> (correct me if I'm wrong on this). I think it would be better
> to solve this with

Well, I definitely think there are other packages like openssl, don't 
underestimate the imagination of developpers, althought it shouldn't be 
the majority.

In blastwave, I think at least pine and procmail don't use the standard 
configure, so it would be nicer to be able to handle that case without 
specific tweak.

> 
>>     - I would like to be able to not use the automatic prototype 
>> generation
>> but I didn't see a way to disable your new automatic prototype
>> generation. Could you add this option ?
> 
> This should work out-of-the-box. If the prototype is defined
> in DISTFILES it shouldn't get overridden, so existing Makefiles
> shouldn't break. Generally dynamic prototypes don't need adjustment
> on version updates, so if there is anything missing which you need to
> use them let me know what is missing and we may find a general solution.

The prototype are parts of DISTFILES, have a look at 
http://svn.blastwave.org/trac/browser/csw/branches/dam/automatic-multiarch-compile/lib/openssl/Makefile
They are arch suffixed, maybe that's not correctly handled.

>>     - I still have a problem building openssl, the first time I try to 
>> build the package target after a "gmake clean", I have the following 
>> error:
>>     gmake: *** No rule to make target `package-CSWossl', needed by 
>> `package'.  Stop.
>>     If retry "gmake package" after this error, it works.
>>
>>     You'll find the debug log of the first gmake call in my home: 
>> ~yann/gmake_package.debug.log
>>     You'll find my automatic-multiarch-compile build tree on ra: 
>> ~yann/csw2/automatic-multiarch-compile/
> 
> This seems to be the same issue: dynamic prototypes. I'll make
> a test environment with your openssl-Makefile and try to reproduce
> this.

Thanks.

Yann


From dam at blastwave.org  Mon Mar 24 23:15:52 2008
From: dam at blastwave.org (Dagobert Michelsen)
Date: Mon, 24 Mar 2008 23:15:52 +0100
Subject: [csw-dev] Multi-ISA GAR
In-Reply-To: <47E822E4.6080608@pleiades.fr.eu.org>
References: <20080319100038.C3150@bolthole.com>
	<33F3A8F3-6FA2-4E2E-91D2-6909A9902BCC@blastwave.org>
	<47E805E3.1060705@pleiades.fr.eu.org>
	<F5645699-8B12-4DD0-BA05-FD383BE95C47@blastwave.org>
	<47E822E4.6080608@pleiades.fr.eu.org>
Message-ID: <D85FED99-46EF-49A2-9B60-6C17067EB839@blastwave.org>

Hi Yann,

Am 24.03.2008 um 22:53 schrieb Yann Rouillard:
> Dagobert Michelsen a ?crit :
>> Am 24.03.2008 um 20:49 schrieb Yann Rouillard:
>>> I began to convert openssl to the new multi-ISA system, I have two
>>> comments so far:
>>>
>>>     - the default isa install target is fine for standard  
>>> programs using automatic-multiarch-compile
>>> the classic configure script but will not work for everything (and
>>>
>>> indeed not for openssl).
>>>     I added a new INSTALL_EXTRA_ISA_SCRIPTS to allow to redefine  
>>> the target
>>> used: see http://svn.blastwave.org/trac/changeset/1994
>>
>> This is a good solution for openssl, but I don't believe
>> that there are any more packages built in the same way
>> (correct me if I'm wrong on this). I think it would be better
>> to solve this with
>
> Well, I definitely think there are other packages like openssl,  
> don't underestimate the imagination of developpers, althought it  
> shouldn't be the majority.
>
> In blastwave, I think at least pine and procmail don't use the  
> standard configure, so it would be nicer to be able to handle that  
> case without specific tweak.

I am not against flexibility here. What I don't like
is the special meaning of ISA_DEFAULT here. You have
a custom installer, which you could call simply by
overriding the generic installer in your Makefile:

   install-isa-pentium_pro install-isa-amd64: install-isa
     $(DONADA)

If there really is need for the special case I'd like
this to be handled in a more general way working for
all ISAs, not only ISA_DEFAULT.


Best regards

   -- Dago

From dam at blastwave.org  Tue Mar 25 09:25:30 2008
From: dam at blastwave.org (Dagobert Michelsen)
Date: Tue, 25 Mar 2008 09:25:30 +0100
Subject: [csw-dev] Multi-ISA GAR
In-Reply-To: <47E822E4.6080608@pleiades.fr.eu.org>
References: <20080319100038.C3150@bolthole.com>
	<33F3A8F3-6FA2-4E2E-91D2-6909A9902BCC@blastwave.org>
	<47E805E3.1060705@pleiades.fr.eu.org>
	<F5645699-8B12-4DD0-BA05-FD383BE95C47@blastwave.org>
	<47E822E4.6080608@pleiades.fr.eu.org>
Message-ID: <87AF833D-9833-4B49-B832-043C9B94FFB2@blastwave.org>

Hi Yann,

Am 24.03.2008 um 22:53 schrieb Yann Rouillard:
>>>     I added a new INSTALL_EXTRA_ISA_SCRIPTS to allow to redefine  
>>> the target
>>> used: see http://svn.blastwave.org/trac/changeset/1994
>> This is a good solution for openssl, but I don't believe
>> that there are any more packages built in the same way
>> (correct me if I'm wrong on this). I think it would be better
>> to solve this with
>
> Well, I definitely think there are other packages like openssl,  
> don't underestimate the imagination of developpers, althought it  
> shouldn't be the majority.
>
> In blastwave, I think at least pine and procmail don't use the  
> standard configure, so it would be nicer to be able to handle that  
> case without specific tweak.

Now I see what you are going to accomplish: you
use the install-part of the Makefile _only_ for
the default ISA and copy in everything else manually.
That is somewhat different from what I had envisioned
but nonetheless useful. James again uses the install
differently: he has separate dirs for DESTDIR-installation
and copies the files together as last step. This may be
elegantly combinable with your approach. Please
continue while I give this some more thought :-)

Best regards

   -- Dago


From dam at blastwave.org  Tue Mar 25 22:00:44 2008
From: dam at blastwave.org (Dagobert Michelsen)
Date: Tue, 25 Mar 2008 22:00:44 +0100
Subject: [csw-dev] Multi-ISA GAR
In-Reply-To: <47E822E4.6080608@pleiades.fr.eu.org>
References: <20080319100038.C3150@bolthole.com>
	<33F3A8F3-6FA2-4E2E-91D2-6909A9902BCC@blastwave.org>
	<47E805E3.1060705@pleiades.fr.eu.org>
	<F5645699-8B12-4DD0-BA05-FD383BE95C47@blastwave.org>
	<47E822E4.6080608@pleiades.fr.eu.org>
Message-ID: <EA0A9D34-86ED-4417-A6AE-9C2D905AE2B3@blastwave.org>

Hi Yann,

Am 24.03.2008 um 22:53 schrieb Yann Rouillard:
> Dagobert Michelsen a ?crit :
>>>     I added a new INSTALL_EXTRA_ISA_SCRIPTS to allow to redefine  
>>> the target
>>> used: see http://svn.blastwave.org/trac/changeset/1994
>>>
>> This is a good solution for openssl, but I don't believe
>> that there are any more packages built in the same way
>> (correct me if I'm wrong on this). I think it would be better
>> to solve this with
>
> Well, I definitely think there are other packages like openssl,  
> don't underestimate the imagination of developpers, althought it  
> shouldn't be the majority.

After some more thinking I think this would be a good idea:
- configure for each ISA to the same standard locations
   (e. g. bindir without $ISA suffix)
- install each ISA to a different, ISA-specific tree
- user-settable scripts allow merging of these trees
   verbatim, select specific binaries or move them to
   $ISA subdirs

I'll may have some time next week to prepare something.

>>>     - I would like to be able to not use the automatic prototype  
>>> generation
>>> but I didn't see a way to disable your new automatic prototype
>>> generation. Could you add this option ?
>>
>> This should work out-of-the-box. If the prototype is defined
>> in DISTFILES it shouldn't get overridden, so existing Makefiles
>> shouldn't break. Generally dynamic prototypes don't need adjustment
>> on version updates, so if there is anything missing which you need to
>> use them let me know what is missing and we may find a general  
>> solution.
>
> The prototype are parts of DISTFILES, have a look at http:// 
> svn.blastwave.org/trac/browser/csw/branches/dam/automatic-multiarch- 
> compile/lib/openssl/Makefile
> They are arch suffixed, maybe that's not correctly handled.

The dependencies in gar.pkg.mk were too strict, so
your files got overwritten each time after install. This
should be fixed in r2023:
   <http://svn.blastwave.org/trac/changeset/2023>

>>>     - I still have a problem building openssl, the first time I  
>>> try to build the package target after a "gmake clean", I have the  
>>> following error:
>>>     gmake: *** No rule to make target `package-CSWossl', needed  
>>> by `package'.  Stop.
>>>     If retry "gmake package" after this error, it works.
>>>
>>>     You'll find the debug log of the first gmake call in my home:  
>>> ~yann/gmake_package.debug.log
>>>     You'll find my automatic-multiarch-compile build tree on ra:  
>>> ~yann/csw2/automatic-multiarch-compile/
>>
>> This seems to be the same issue: dynamic prototypes. I'll make
>> a test environment with your openssl-Makefile and try to reproduce
>> this.
>
> Thanks.

I couldn't verify this although I copied your tree. Please
check again with r2023. If the error is still there I see
what I can do on your tree directly.


Best regards

   -- Dago

From aubreylee at gmail.com  Sun Mar 30 11:44:33 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Sun, 30 Mar 2008 17:44:33 +0800
Subject: [csw-dev] gcc issue
Message-ID: <6d6a94c50803300244u69318b00v68aa5d4fe9d793f4@mail.gmail.com>

Hi list,

I saw this sentence in the build standard:
<Our standard compiler is the sun compiler(s),
which should be used in preference to gcc,
unless the software only compiles with gcc.>

Unfortunately, the package I want to maintain is only
compiled with gcc. But I couldn't found gcc on thor.blastwave.org.

Please tell me what I should do.

Many thanks,
-Aubrey


From william at wbonnet.net  Sun Mar 30 12:36:24 2008
From: william at wbonnet.net (William Bonnet)
Date: Sun, 30 Mar 2008 12:36:24 +0200
Subject: [csw-dev] gcc issue
In-Reply-To: <6d6a94c50803300244u69318b00v68aa5d4fe9d793f4@mail.gmail.com>
References: <6d6a94c50803300244u69318b00v68aa5d4fe9d793f4@mail.gmail.com>
Message-ID: <47EF6D28.7090204@wbonnet.net>

Hi
> Unfortunately, the package I want to maintain is only
> compiled with gcc. But I couldn't found gcc on thor.blastwave.org.
>
> Please tell me what I should do.
>   
gcc is in /opt/csw/gcc3/bin

cheers

-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.blastwave.org An OpenSolaris Community Site
http://www.guses.org     French speaking Solaris User Group




From aubreylee at gmail.com  Sun Mar 30 12:50:46 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Sun, 30 Mar 2008 18:50:46 +0800
Subject: [csw-dev] gcc issue
In-Reply-To: <47EF6D28.7090204@wbonnet.net>
References: <6d6a94c50803300244u69318b00v68aa5d4fe9d793f4@mail.gmail.com>
	<47EF6D28.7090204@wbonnet.net>
Message-ID: <6d6a94c50803300350h21d373a9p467e45c621189967@mail.gmail.com>

Hi William,

On Sun, Mar 30, 2008 at 6:36 PM, William Bonnet <william at wbonnet.net> wrote:
> Hi
>
> > Unfortunately, the package I want to maintain is only
>  > compiled with gcc. But I couldn't found gcc on thor.blastwave.org.
>  >
>  > Please tell me what I should do.
>  >
>  gcc is in /opt/csw/gcc3/bin
>
Thanks, I thought gcc is in /usr/sfw/bin, I got it now, ;-)

-Aubrey


From aubreylee at gmail.com  Sun Mar 30 13:29:30 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Sun, 30 Mar 2008 19:29:30 +0800
Subject: [csw-dev] getopt.h missing on solaris8
Message-ID: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>

I found thor.blastwave.org is a solaris8 system,

bash-2.03$ cat /etc/release
                       Solaris 8 2/02 s28x_u7wos_08a INTEL
           Copyright 2002 Sun Microsystems, Inc.  All Rights Reserved.
                           Assembled 18 December 2001

And hope I didn't miss it again, getopt.h and its lib is missing on this box.
what is the right way to fix this issue? it's already there on my nevada system.

Thanks,
-Aubrey


From aubreylee at gmail.com  Sun Mar 30 13:46:52 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Sun, 30 Mar 2008 19:46:52 +0800
Subject: [csw-dev] checkpkg problem
Message-ID: <6d6a94c50803300446l75ee5120h8ef2db4af4e11f55@mail.gmail.com>

I got the following error on my side.

Running checkpkg on pkgfile
Examining pmtools-20071116-SunOS5.11-i386-CSW.pkg
Looking for bad strings...
Extracting files for more detailed inspection...
CSWpmtools/pkginfo
63+0 records in
63+0 records out
2 blocks
ERROR: filename should start with -20071116-
20071116 pmtools-20071116-SunOS5.11-i386-CSW.pkg
ERROR: failed check.

here is my pkginfo
========
PKG=CSWpmtools
NAME=pmtools
VERSION=20071116
CATEGORY=application
VENDOR=http://www.lesswatts.org/patches/linux_acpi/ packaged for CSW
by Aubrey Li
HOTLINE=http://www.blastwave.org/bugtrack/
========

The problem is the following line in /opt/csw/bin/checkpkg
====
software=`sed -n 's/^NAME=\([^ -]*\) -.*$/\1/p' $TMPFILE`
====
software is assigned NULL by sed.

I don't think my pkginfo file has problem, so I double checked on the shell:
------------------------
[aubrey-sz330p at pmtools]sed -n 's/^VERSION=//p' pkginfo
20071116
------------------------
VERSION is no problem.

-----------------------
#sed -n 's/^NAME=\([^ -]*\) -.*$/\1/p' pkginfo
-----------------------
NAME return nothing.

Is this a problem in checkpkg? or I did something wrong?

I'm really appreciated any hints!

Thanks,
-Aubrey


From Murray.Jensen at csiro.au  Sun Mar 30 15:29:15 2008
From: Murray.Jensen at csiro.au (Murray.Jensen at csiro.au)
Date: Mon, 31 Mar 2008 00:29:15 +1100
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: Your message of "Sun, 30 Mar 2008 19:29:30 +0800"
	<6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com> 
Message-ID: <22844.1206883755@gerd>

On Sun, 30 Mar 2008 19:29:30 +0800, "Aubrey Li" writes:
>getopt.h and its lib is missing on this box.
>what is the right way to fix this issue?

Solaris 8 doesn't have getopt.h - or the gnu getopt* functions in libc.

Grab the source from gnulib and link it in with your application ...

	http://www.gnu.org/software/gnulib/

or direct to getopt module ...

	http://www.gnu.org/software/gnulib/MODULES.html#module=getopt

There are links for each of the following files

	getopt.in.h
	getopt.c
	getopt1.c
	getopt_int.h

but the first one is bogus - it should be getopt.h ... click on the
link and when you get the error page, go into the address bar and
edit the url such that getopt.in.h is changed to getopt.h.

I think those four files will be enough ... you will have to figure
out how to link them in - usually edit a Makefile, Makefile.in or
Makefile.am file.

This has been discussed in the maintainers list before - but its not
your fault you didnt find it - the archives arent easily searchable
(unless someone else has a searchable interface?). Cheers!
								Murray...
--
Murray Jensen, CSIRO Materials Science and Engineering Phone: +61 3 9545 2075
Private Bag 33, Clayton South 3169, Australia            Fax: +61 3 9544 1128
Internet: Murray.Jensen at csiro.au

To the extent permitted by law, CSIRO does not represent, warrant and/or
guarantee that the integrity of this communication has been maintained or
that the communication is free of errors, virus, interception or interference.

The information contained in this e-mail may be confidential or privileged.
Any unauthorised use or disclosure is prohibited. If you have received this
e-mail in error, please delete it immediately and notify Murray Jensen on
+61 3 9545 2075. Thank you.


From Murray.Jensen at csiro.au  Sun Mar 30 15:32:10 2008
From: Murray.Jensen at csiro.au (Murray.Jensen at csiro.au)
Date: Mon, 31 Mar 2008 00:32:10 +1100
Subject: [csw-dev] checkpkg problem
In-Reply-To: Your message of "Sun, 30 Mar 2008 19:46:52 +0800"
	<6d6a94c50803300446l75ee5120h8ef2db4af4e11f55@mail.gmail.com> 
Message-ID: <22901.1206883930@gerd>

On Sun, 30 Mar 2008 19:46:52 +0800, "Aubrey Li" writes:
>NAME=pmtools

NAME must have the "real" package name (i.e. not the system V CSW... name)
followed by " - " and a short description (the sed pattern should have
given it away). Cheers!
								Murray...
--
Murray Jensen, CSIRO Materials Science and Engineering Phone: +61 3 9545 2075
Private Bag 33, Clayton South 3169, Australia            Fax: +61 3 9544 1128
Internet: Murray.Jensen at csiro.au

To the extent permitted by law, CSIRO does not represent, warrant and/or
guarantee that the integrity of this communication has been maintained or
that the communication is free of errors, virus, interception or interference.

The information contained in this e-mail may be confidential or privileged.
Any unauthorised use or disclosure is prohibited. If you have received this
e-mail in error, please delete it immediately and notify Murray Jensen on
+61 3 9545 2075. Thank you.


From aubreylee at gmail.com  Sun Mar 30 17:09:33 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Sun, 30 Mar 2008 23:09:33 +0800
Subject: [csw-dev] checkpkg problem
In-Reply-To: <22901.1206883930@gerd>
References: <6d6a94c50803300446l75ee5120h8ef2db4af4e11f55@mail.gmail.com>
	<22901.1206883930@gerd>
Message-ID: <6d6a94c50803300809v424b8c7bx7f1cf574aacf6252@mail.gmail.com>

On Sun, Mar 30, 2008 at 9:32 PM,  <Murray.Jensen at csiro.au> wrote:
> On Sun, 30 Mar 2008 19:46:52 +0800, "Aubrey Li" writes:
>  >NAME=pmtools
>
>  NAME must have the "real" package name (i.e. not the system V CSW... name)
>  followed by " - " and a short description (the sed pattern should have
>  given it away). Cheers!
>                                                                 Murray...
Many thanks, I got it.

-Aubrey


From aubreylee at gmail.com  Sun Mar 30 17:16:32 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Sun, 30 Mar 2008 23:16:32 +0800
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <22844.1206883755@gerd>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<22844.1206883755@gerd>
Message-ID: <6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>

On Sun, Mar 30, 2008 at 9:29 PM,  <Murray.Jensen at csiro.au> wrote:
> On Sun, 30 Mar 2008 19:29:30 +0800, "Aubrey Li" writes:
>  >getopt.h and its lib is missing on this box.
>  >what is the right way to fix this issue?
>
>  Solaris 8 doesn't have getopt.h - or the gnu getopt* functions in libc.
>
>  Grab the source from gnulib and link it in with your application ...
>
>         http://www.gnu.org/software/gnulib/
>
>  or direct to getopt module ...
>
>         http://www.gnu.org/software/gnulib/MODULES.html#module=getopt
>
>  There are links for each of the following files
>
>         getopt.in.h
>         getopt.c
>         getopt1.c
>         getopt_int.h
>
>  but the first one is bogus - it should be getopt.h ... click on the
>  link and when you get the error page, go into the address bar and
>  edit the url such that getopt.in.h is changed to getopt.h.
>
>  I think those four files will be enough ... you will have to figure
>  out how to link them in - usually edit a Makefile, Makefile.in or
>  Makefile.am file.
>
>  This has been discussed in the maintainers list before - but its not
>  your fault you didnt find it - the archives arent easily searchable
>  (unless someone else has a searchable interface?). Cheers!
>                                                                 Murray...

Thanks Murray, I'll have a try.
Just be curious, why don't we upgrade our system to solaris 10?

-Aubrey


From Joerg.Schilling at fokus.fraunhofer.de  Sun Mar 30 17:28:47 2008
From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling)
Date: Sun, 30 Mar 2008 17:28:47 +0200
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
Message-ID: <47efb1af.1k1w0/oKWDynZgCC%Joerg.Schilling@fokus.fraunhofer.de>

"Aubrey Li" <aubreylee at gmail.com> wrote:

> I found thor.blastwave.org is a solaris8 system,
>
> bash-2.03$ cat /etc/release
>                        Solaris 8 2/02 s28x_u7wos_08a INTEL
>            Copyright 2002 Sun Microsystems, Inc.  All Rights Reserved.
>                            Assembled 18 December 2001
>
> And hope I didn't miss it again, getopt.h and its lib is missing on this box.
> what is the right way to fix this issue? it's already there on my nevada system.

getopt.h does not officieally exist, it is a private file written by RMS.

J?rg

-- 
 EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin
       js at cs.tu-berlin.de                (uni)  
       schilling at fokus.fraunhofer.de     (work) Blog: http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily


From dclarke at blastwave.org  Sun Mar 30 17:35:25 2008
From: dclarke at blastwave.org (Dennis Clarke)
Date: Sun, 30 Mar 2008 11:35:25 -0400 (EDT)
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<22844.1206883755@gerd>
	<6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>
Message-ID: <2056.72.39.133.97.1206891325.squirrel@mail.blastwave.org>


> On Sun, Mar 30, 2008 at 9:29 PM,  <Murray.Jensen at csiro.au> wrote:
>> On Sun, 30 Mar 2008 19:29:30 +0800, "Aubrey Li" writes:
>>  >getopt.h and its lib is missing on this box.
>>  >what is the right way to fix this issue?
>>
>>  Solaris 8 doesn't have getopt.h - or the gnu getopt* functions in libc.
>>
>>  Grab the source from gnulib and link it in with your application ...
>>
>>         http://www.gnu.org/software/gnulib/
>>
>>  or direct to getopt module ...
>>
>>         http://www.gnu.org/software/gnulib/MODULES.html#module=getopt
>>
>>  There are links for each of the following files
>>
>>         getopt.in.h
>>         getopt.c
>>         getopt1.c
>>         getopt_int.h
>>
>>  but the first one is bogus - it should be getopt.h ... click on the
>>  link and when you get the error page, go into the address bar and
>>  edit the url such that getopt.in.h is changed to getopt.h.
>>
>>  I think those four files will be enough ... you will have to figure
>>  out how to link them in - usually edit a Makefile, Makefile.in or
>>  Makefile.am file.
>>
>>  This has been discussed in the maintainers list before - but its not
>>  your fault you didnt find it - the archives arent easily searchable
>>  (unless someone else has a searchable interface?). Cheers!
>>                                                                 Murray...
>
> Thanks Murray, I'll have a try.
> Just be curious, why don't we upgrade our system to solaris 10?

Solaris 8 is scheduled to be dropped entirely into the "time and material"
support class by Sun for Q12009.  That is March 2009.  I hope to have all of
the Solaris 10 based infrastructure up to speed before then.

Dennis



From dclarke at blastwave.org  Sun Mar 30 17:38:29 2008
From: dclarke at blastwave.org (Dennis Clarke)
Date: Sun, 30 Mar 2008 11:38:29 -0400 (EDT)
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <47efb1af.1k1w0/oKWDynZgCC%Joerg.Schilling@fokus.fraunhofer.de>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<47efb1af.1k1w0/oKWDynZgCC%Joerg.Schilling@fokus.fraunhofer.de>
Message-ID: <2059.72.39.133.97.1206891509.squirrel@mail.blastwave.org>


> "Aubrey Li" <aubreylee at gmail.com> wrote:
>
>> I found thor.blastwave.org is a solaris8 system,
>>
>> bash-2.03$ cat /etc/release
>>                        Solaris 8 2/02 s28x_u7wos_08a INTEL
>>            Copyright 2002 Sun Microsystems, Inc.  All Rights Reserved.
>>                            Assembled 18 December 2001
>>
>> And hope I didn't miss it again, getopt.h and its lib is missing on this
>> box.
>> what is the right way to fix this issue? it's already there on my nevada
>> system.
>
> getopt.h does not officieally exist, it is a private file written by RMS.
>
> J?rg

In the category "J?rg will know for sure"  :-)

What is the correct way to do things? Officially ?

Dennis



From Joerg.Schilling at fokus.fraunhofer.de  Sun Mar 30 17:45:20 2008
From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling)
Date: Sun, 30 Mar 2008 17:45:20 +0200
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<22844.1206883755@gerd>
	<6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>
Message-ID: <47efb590.GEPvZ3W62E/R68/e%Joerg.Schilling@fokus.fraunhofer.de>

"Aubrey Li" <aubreylee at gmail.com> wrote:

> >  This has been discussed in the maintainers list before - but its not
> >  your fault you didnt find it - the archives arent easily searchable
> >  (unless someone else has a searchable interface?). Cheers!
> >                                                                 Murray...
>
> Thanks Murray, I'll have a try.
> Just be curious, why don't we upgrade our system to solaris 10?

Sources that asume the existence of a GNU/RMS private include file cannot
be called portable. You should inform the authors.

J?rg

-- 
 EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin
       js at cs.tu-berlin.de                (uni)  
       schilling at fokus.fraunhofer.de     (work) Blog: http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily


From Joerg.Schilling at fokus.fraunhofer.de  Sun Mar 30 17:50:15 2008
From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling)
Date: Sun, 30 Mar 2008 17:50:15 +0200
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <2059.72.39.133.97.1206891509.squirrel@mail.blastwave.org>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<47efb1af.1k1w0/oKWDynZgCC%Joerg.Schilling@fokus.fraunhofer.de>
	<2059.72.39.133.97.1206891509.squirrel@mail.blastwave.org>
Message-ID: <47efb6b7.COsZ6N/5WPc3uyjH%Joerg.Schilling@fokus.fraunhofer.de>

"Dennis Clarke" <dclarke at blastwave.org> wrote:

> > getopt.h does not officieally exist, it is a private file written by RMS.
> >
> > J?rg
>
> In the category "J?rg will know for sure"  :-)
>
> What is the correct way to do things? Officially ?

The getopt() prototype is in unistd.h

Most GNU influenced software tests for getopt_long() but forgets
about the fact that getopt.h is even more non-standard.


http://www.opengroup.org/onlinepubs/009695399/functions/getopt.html

is the official standard.

J?rg

-- 
 EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin
       js at cs.tu-berlin.de                (uni)  
       schilling at fokus.fraunhofer.de     (work) Blog: http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily


From aubreylee at gmail.com  Sun Mar 30 17:57:23 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Sun, 30 Mar 2008 23:57:23 +0800
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <2056.72.39.133.97.1206891325.squirrel@mail.blastwave.org>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<22844.1206883755@gerd>
	<6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>
	<2056.72.39.133.97.1206891325.squirrel@mail.blastwave.org>
Message-ID: <6d6a94c50803300857o302c8769i3668606ae7c0cf2e@mail.gmail.com>

On Sun, Mar 30, 2008 at 11:35 PM, Dennis Clarke <dclarke at blastwave.org> wrote:
>
>
>  > On Sun, Mar 30, 2008 at 9:29 PM,  <Murray.Jensen at csiro.au> wrote:
>  >> On Sun, 30 Mar 2008 19:29:30 +0800, "Aubrey Li" writes:
>  >>  >getopt.h and its lib is missing on this box.
>  >>  >what is the right way to fix this issue?
>  >>
>  >>  Solaris 8 doesn't have getopt.h - or the gnu getopt* functions in libc.
>  >>
>  >>  Grab the source from gnulib and link it in with your application ...
>  >>
>  >>         http://www.gnu.org/software/gnulib/
>  >>
>  >>  or direct to getopt module ...
>  >>
>  >>         http://www.gnu.org/software/gnulib/MODULES.html#module=getopt
>  >>
>  >>  There are links for each of the following files
>  >>
>  >>         getopt.in.h
>  >>         getopt.c
>  >>         getopt1.c
>  >>         getopt_int.h
>  >>
>  >>  but the first one is bogus - it should be getopt.h ... click on the
>  >>  link and when you get the error page, go into the address bar and
>  >>  edit the url such that getopt.in.h is changed to getopt.h.
>  >>
>  >>  I think those four files will be enough ... you will have to figure
>  >>  out how to link them in - usually edit a Makefile, Makefile.in or
>  >>  Makefile.am file.
>  >>
>  >>  This has been discussed in the maintainers list before - but its not
>  >>  your fault you didnt find it - the archives arent easily searchable
>  >>  (unless someone else has a searchable interface?). Cheers!
>  >>                                                                 Murray...
>  >
>  > Thanks Murray, I'll have a try.
>  > Just be curious, why don't we upgrade our system to solaris 10?
>
>  Solaris 8 is scheduled to be dropped entirely into the "time and material"
>  support class by Sun for Q12009.  That is March 2009.  I hope to have all of
>  the Solaris 10 based infrastructure up to speed before then.
>
>  Dennis
>

Of course I can't wait then, my account will be invalid in one month, ;-)

-Aubrey


From dclarke at blastwave.org  Sun Mar 30 18:22:05 2008
From: dclarke at blastwave.org (Dennis Clarke)
Date: Sun, 30 Mar 2008 12:22:05 -0400 (EDT)
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <47efb6b7.COsZ6N/5WPc3uyjH%Joerg.Schilling@fokus.fraunhofer.de>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<47efb1af.1k1w0/oKWDynZgCC%Joerg.Schilling@fokus.fraunhofer.de>
	<2059.72.39.133.97.1206891509.squirrel@mail.blastwave.org>
	<47efb6b7.COsZ6N/5WPc3uyjH%Joerg.Schilling@fokus.fraunhofer.de>
Message-ID: <2246.72.39.133.97.1206894125.squirrel@mail.blastwave.org>


> "Dennis Clarke" <dclarke at blastwave.org> wrote:
>
>> > getopt.h does not officieally exist, it is a private file written by
>> RMS.
>> >
>> > J?rg
>>
>> In the category "J?rg will know for sure"  :-)
>>
>> What is the correct way to do things? Officially ?
>
> The getopt() prototype is in unistd.h
>
> Most GNU influenced software tests for getopt_long() but forgets
> about the fact that getopt.h is even more non-standard.
>
>
> http://www.opengroup.org/onlinepubs/009695399/functions/getopt.html
>
> is the official standard.
>

I knew that all the standards of value are at the OpenGroup but was not sure
of the link. So I sat back and waited and sure enough .. you knew :-)

So the issue is with the upstream coders that target Linux users and that is
becoming a defacto way of things these days.

Dennis



From dclarke at blastwave.org  Sun Mar 30 18:29:23 2008
From: dclarke at blastwave.org (Dennis Clarke)
Date: Sun, 30 Mar 2008 12:29:23 -0400 (EDT)
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <6d6a94c50803300857o302c8769i3668606ae7c0cf2e@mail.gmail.com>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<22844.1206883755@gerd>
	<6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>
	<2056.72.39.133.97.1206891325.squirrel@mail.blastwave.org>
	<6d6a94c50803300857o302c8769i3668606ae7c0cf2e@mail.gmail.com>
Message-ID: <2317.72.39.133.97.1206894563.squirrel@mail.blastwave.org>


> On Sun, Mar 30, 2008 at 11:35 PM, Dennis Clarke <dclarke at blastwave.org>
> wrote:
>>
>>
>>  > On Sun, Mar 30, 2008 at 9:29 PM,  <Murray.Jensen at csiro.au> wrote:
>>  >> On Sun, 30 Mar 2008 19:29:30 +0800, "Aubrey Li" writes:
>>  >>  >getopt.h and its lib is missing on this box.
>>  >>  >what is the right way to fix this issue?
>>  >>
>>  >>  Solaris 8 doesn't have getopt.h - or the gnu getopt* functions in
>> libc.
>>  >>
>>  >>  Grab the source from gnulib and link it in with your application ...
>>  >>
>>  >>         http://www.gnu.org/software/gnulib/
>>  >>
>>  >>  or direct to getopt module ...
>>  >>
>>  >>         http://www.gnu.org/software/gnulib/MODULES.html#module=getopt
>>  >>
>>  >>  There are links for each of the following files
>>  >>
>>  >>         getopt.in.h
>>  >>         getopt.c
>>  >>         getopt1.c
>>  >>         getopt_int.h
>>  >>
>>  >>  but the first one is bogus - it should be getopt.h ... click on the
>>  >>  link and when you get the error page, go into the address bar and
>>  >>  edit the url such that getopt.in.h is changed to getopt.h.
>>  >>
>>  >>  I think those four files will be enough ... you will have to figure
>>  >>  out how to link them in - usually edit a Makefile, Makefile.in or
>>  >>  Makefile.am file.
>>  >>
>>  >>  This has been discussed in the maintainers list before - but its not
>>  >>  your fault you didnt find it - the archives arent easily searchable
>>  >>  (unless someone else has a searchable interface?). Cheers!
>>  >>                                                                 Murray...
>>  >
>>  > Thanks Murray, I'll have a try.
>>  > Just be curious, why don't we upgrade our system to solaris 10?
>>
>>  Solaris 8 is scheduled to be dropped entirely into the "time and
>> material"
>>  support class by Sun for Q12009.  That is March 2009.  I hope to have all
>> of
>>  the Solaris 10 based infrastructure up to speed before then.
>>
>>  Dennis
>>
>
> Of course I can't wait then, my account will be invalid in one month, ;-)

Say what ?  What account ?

Dennis



From aubreylee at gmail.com  Sun Mar 30 19:41:01 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Mon, 31 Mar 2008 01:41:01 +0800
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <2317.72.39.133.97.1206894563.squirrel@mail.blastwave.org>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<22844.1206883755@gerd>
	<6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>
	<2056.72.39.133.97.1206891325.squirrel@mail.blastwave.org>
	<6d6a94c50803300857o302c8769i3668606ae7c0cf2e@mail.gmail.com>
	<2317.72.39.133.97.1206894563.squirrel@mail.blastwave.org>
Message-ID: <6d6a94c50803301041p1a0e25dave254093344c58323@mail.gmail.com>

On Mon, Mar 31, 2008 at 12:29 AM, Dennis Clarke <dclarke at blastwave.org> wrote:
>
>
>  > On Sun, Mar 30, 2008 at 11:35 PM, Dennis Clarke <dclarke at blastwave.org>
>  > wrote:
>  >>
>  >>
>  >>  > On Sun, Mar 30, 2008 at 9:29 PM,  <Murray.Jensen at csiro.au> wrote:
>  >>  >> On Sun, 30 Mar 2008 19:29:30 +0800, "Aubrey Li" writes:
>  >>  >>  >getopt.h and its lib is missing on this box.
>  >>  >>  >what is the right way to fix this issue?
>  >>  >>
>  >>  >>  Solaris 8 doesn't have getopt.h - or the gnu getopt* functions in
>  >> libc.
>  >>  >>
>  >>  >>  Grab the source from gnulib and link it in with your application ...
>  >>  >>
>  >>  >>         http://www.gnu.org/software/gnulib/
>  >>  >>
>  >>  >>  or direct to getopt module ...
>  >>  >>
>  >>  >>         http://www.gnu.org/software/gnulib/MODULES.html#module=getopt
>  >>  >>
>  >>  >>  There are links for each of the following files
>  >>  >>
>  >>  >>         getopt.in.h
>  >>  >>         getopt.c
>  >>  >>         getopt1.c
>  >>  >>         getopt_int.h
>  >>  >>
>  >>  >>  but the first one is bogus - it should be getopt.h ... click on the
>  >>  >>  link and when you get the error page, go into the address bar and
>  >>  >>  edit the url such that getopt.in.h is changed to getopt.h.
>  >>  >>
>  >>  >>  I think those four files will be enough ... you will have to figure
>  >>  >>  out how to link them in - usually edit a Makefile, Makefile.in or
>  >>  >>  Makefile.am file.
>  >>  >>
>  >>  >>  This has been discussed in the maintainers list before - but its not
>  >>  >>  your fault you didnt find it - the archives arent easily searchable
>  >>  >>  (unless someone else has a searchable interface?). Cheers!
>  >>  >>                                                                 Murray...
>  >>  >
>  >>  > Thanks Murray, I'll have a try.
>  >>  > Just be curious, why don't we upgrade our system to solaris 10?
>  >>
>  >>  Solaris 8 is scheduled to be dropped entirely into the "time and
>  >> material"
>  >>  support class by Sun for Q12009.  That is March 2009.  I hope to have all
>  >> of
>  >>  the Solaris 10 based infrastructure up to speed before then.
>  >>
>  >>  Dennis
>  >>
>  >
>  > Of course I can't wait then, my account will be invalid in one month, ;-)
>
>  Say what ?  What account ?
For new maintainer, I have 30 days to create my first package, or my
account will expire.

Anyway, I made a simple link getopt staff in my application, and my
first package
was born successfully, ;-)

Thanks,
-Aubrey


From aubreylee at gmail.com  Mon Mar 31 07:26:18 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Mon, 31 Mar 2008 13:26:18 +0800
Subject: [csw-dev] checkpkg problem
In-Reply-To: <6d6a94c50803300809v424b8c7bx7f1cf574aacf6252@mail.gmail.com>
References: <6d6a94c50803300446l75ee5120h8ef2db4af4e11f55@mail.gmail.com>
	<22901.1206883930@gerd>
	<6d6a94c50803300809v424b8c7bx7f1cf574aacf6252@mail.gmail.com>
Message-ID: <6d6a94c50803302226m74bd433djf53f2bd44ad8d9c8@mail.gmail.com>

On Sun, Mar 30, 2008 at 11:09 PM, Aubrey Li <aubreylee at gmail.com> wrote:
> On Sun, Mar 30, 2008 at 9:32 PM,  <Murray.Jensen at csiro.au> wrote:
>  > On Sun, 30 Mar 2008 19:46:52 +0800, "Aubrey Li" writes:
>  >  >NAME=pmtools
>  >
>  >  NAME must have the "real" package name (i.e. not the system V CSW... name)
>  >  followed by " - " and a short description (the sed pattern should have
>  >  given it away). Cheers!
>  >                                                                 Murray...

I got another check error - cpio error, see below please.
the package is ok except this error.
any hints?

Thanks,
-Aubrey
====================================
[aubrey-sz330p at pmtools]createpkg -b
/export/home/aubrey/pkg-home/build/pmtools-20071116/cswstage/opt/csw
i copyright
## Building pkgmap from package prototype file.
## Processing pkginfo file.
WARNING: parameter <PSTAMP> set to "aubrey-sz330p20080331131732"
WARNING: parameter <CLASSES> set to "none"
## Attempting to volumize 5 entries in pkgmap.
part  1 -- 108 blocks, 15 entries
## Packaging one part.
/var/spool/pkg/CSWpmtools/pkgmap
/var/spool/pkg/CSWpmtools/pkginfo
/var/spool/pkg/CSWpmtools/reloc/bin/acpidump
/var/spool/pkg/CSWpmtools/reloc/bin/acpitbl
/var/spool/pkg/CSWpmtools/reloc/bin/acpixtract
/var/spool/pkg/CSWpmtools/reloc/bin/madt
/var/spool/pkg/CSWpmtools/install/copyright
/var/spool/pkg/CSWpmtools/install/depend
## Validating control scripts.
## Packaging complete.
Transferring <CSWpmtools> package instance
Conversion complete. Cleaning up in /var/spool/pkg

Running checkpkg on pkgfile
Examining pmtools-20071116,REV=2008.03.30-SunOS5.11-i386-CSW.pkg
Looking for bad strings...
Extracting files for more detailed inspection...
CSWpmtools/pkginfo
2 blocks
Info for pkg CSWpmtools
software= 'pmtools'
arch= 'i386'
desc='ACPI utilities'
pkgname='CSWpmtools'
version='20071116,REV=2008.03.30'
maintainer name='Aubrey Li'
maint email='aubrey at blastwave.org'
hotline='http://www.blastwave.org/bugtrack/'
basedir='/opt/csw'

Extracing pkg for examination of files...
cpio: Impossible header type.         <============= here
1 errors
retrying extract with different archive offset...
pkginfo
pkgmap
install
install/copyright
install/depend
reloc
reloc/bin
reloc/bin/acpidump
reloc/bin/acpitbl
reloc/bin/acpixtract
reloc/bin/madt
89 blocks
Checking uninstalled directory format package <CSWpmtools> from
</tmp/dissect.102987>
## Checking control scripts.
## Checking package objects.
## Checking is complete.
/usr/sbin/pkgchk passed.
libraries used are:
libc.so.1
cross-referencing with depend file (May take a while)

Examining 'depend' file
system      CSWcommon common - common files and dirs for CSW packages
system      SUNWcsl Core Solaris, (Shared Libs)
system      SUNWdpl Developer Profiled Libraries
SUGGESTION: you may want to add some or all of the following as depends:
   (Feel free to ignore SUNW or SPRO packages)
> SUNWcslr

Checking relocation ability...
trivial check passed

packaging of files under
/export/home/aubrey/pkg-home/build/pmtools-20071116/cswstage/opt/csw
complete
gzipping pkgfile


From Joerg.Schilling at fokus.fraunhofer.de  Mon Mar 31 16:47:24 2008
From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling)
Date: Mon, 31 Mar 2008 16:47:24 +0200
Subject: [csw-dev] checkpkg problem
In-Reply-To: <6d6a94c50803302226m74bd433djf53f2bd44ad8d9c8@mail.gmail.com>
References: <6d6a94c50803300446l75ee5120h8ef2db4af4e11f55@mail.gmail.com>
	<22901.1206883930@gerd>
	<6d6a94c50803300809v424b8c7bx7f1cf574aacf6252@mail.gmail.com>
	<6d6a94c50803302226m74bd433djf53f2bd44ad8d9c8@mail.gmail.com>
Message-ID: <47f0f97c.70tVzGx/kXhfRnQB%Joerg.Schilling@fokus.fraunhofer.de>

"Aubrey Li" <aubreylee at gmail.com> wrote:

> On Sun, Mar 30, 2008 at 11:09 PM, Aubrey Li <aubreylee at gmail.com> wrote:
> > On Sun, Mar 30, 2008 at 9:32 PM,  <Murray.Jensen at csiro.au> wrote:
> >  > On Sun, 30 Mar 2008 19:46:52 +0800, "Aubrey Li" writes:
> >  >  >NAME=pmtools
> >  >
> >  >  NAME must have the "real" package name (i.e. not the system V CSW... name)
> >  >  followed by " - " and a short description (the sed pattern should have
> >  >  given it away). Cheers!
> >  >                                                                 Murray...
>
> I got another check error - cpio error, see below please.
> the package is ok except this error.
> any hints?
>
> desc='ACPI utilities'
> pkgname='CSWpmtools'
> version='20071116,REV=2008.03.30'
> maintainer name='Aubrey Li'
> maint email='aubrey at blastwave.org'
> hotline='http://www.blastwave.org/bugtrack/'
> basedir='/opt/csw'
>
> Extracing pkg for examination of files...
> cpio: Impossible header type.         <============= here

This message is usually seen when Sun's cpio tries to extract a POSIXly correct
TAR archive.

J?rg

-- 
 EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin
       js at cs.tu-berlin.de                (uni)  
       schilling at fokus.fraunhofer.de     (work) Blog: http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily


From phil at bolthole.com  Mon Mar 31 17:33:56 2008
From: phil at bolthole.com (Philip Brown)
Date: Mon, 31 Mar 2008 08:33:56 -0700
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>;
	from aubreylee@gmail.com on Sun, Mar 30, 2008 at 11:16:32PM +0800
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<22844.1206883755@gerd>
	<6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>
Message-ID: <20080331083356.A99037@bolthole.com>

On Sun, Mar 30, 2008 at 11:16:32PM +0800, Aubrey Li wrote:
> Thanks Murray, I'll have a try.
> Just be curious, why don't we upgrade our system to solaris 10?


The mission of the CSW packaging effort can be started as providing
"supported packages of freeware", for "supported versions of solaris".

Solaris 8 is the oldest version of solaris fully supported by sun.
Therefore, all our packages have to support solaris 8 at this time.

This is mentioned with somewhat different wording, at
http://www.blastwave.org/standards/core_principles.html


From aubreylee at gmail.com  Mon Mar 31 18:24:15 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Tue, 1 Apr 2008 00:24:15 +0800
Subject: [csw-dev] checkpkg problem
In-Reply-To: <47f0f97c.70tVzGx/kXhfRnQB%Joerg.Schilling@fokus.fraunhofer.de>
References: <6d6a94c50803300446l75ee5120h8ef2db4af4e11f55@mail.gmail.com>
	<22901.1206883930@gerd>
	<6d6a94c50803300809v424b8c7bx7f1cf574aacf6252@mail.gmail.com>
	<6d6a94c50803302226m74bd433djf53f2bd44ad8d9c8@mail.gmail.com>
	<47f0f97c.70tVzGx/kXhfRnQB%Joerg.Schilling@fokus.fraunhofer.de>
Message-ID: <6d6a94c50803310924n5c02cdffrcf882ef1cb90f1c1@mail.gmail.com>

On Mon, Mar 31, 2008 at 10:47 PM, Joerg Schilling
<Joerg.Schilling at fokus.fraunhofer.de> wrote:
> "Aubrey Li" <aubreylee at gmail.com> wrote:
>
>  > On Sun, Mar 30, 2008 at 11:09 PM, Aubrey Li <aubreylee at gmail.com> wrote:
>  > > On Sun, Mar 30, 2008 at 9:32 PM,  <Murray.Jensen at csiro.au> wrote:
>  > >  > On Sun, 30 Mar 2008 19:46:52 +0800, "Aubrey Li" writes:
>  > >  >  >NAME=pmtools
>  > >  >
>  > >  >  NAME must have the "real" package name (i.e. not the system V CSW... name)
>  > >  >  followed by " - " and a short description (the sed pattern should have
>  > >  >  given it away). Cheers!
>  > >  >                                                                 Murray...
>  >
>  > I got another check error - cpio error, see below please.
>  > the package is ok except this error.
>  > any hints?
>  >
>
> > desc='ACPI utilities'
>  > pkgname='CSWpmtools'
>  > version='20071116,REV=2008.03.30'
>  > maintainer name='Aubrey Li'
>  > maint email='aubrey at blastwave.org'
>  > hotline='http://www.blastwave.org/bugtrack/'
>  > basedir='/opt/csw'
>  >
>  > Extracing pkg for examination of files...
>  > cpio: Impossible header type.         <============= here
>
>  This message is usually seen when Sun's cpio tries to extract a POSIXly correct
>  TAR archive.
>
So do you mean the package has no problem, and we can ignore this
message, can't we?
If we can't ignore this message, how to fix this issue? I just used
the script to create package.
Is there any choice to make a TAR archive which makes Sun's cpio happy?
or don't use Sun's cpio?

Thanks,
-Aubrey


From aubreylee at gmail.com  Mon Mar 31 18:25:36 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Tue, 1 Apr 2008 00:25:36 +0800
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <20080331083356.A99037@bolthole.com>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<22844.1206883755@gerd>
	<6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>
	<20080331083356.A99037@bolthole.com>
Message-ID: <6d6a94c50803310925y5fadef0ak7e54bf3b02f02fec@mail.gmail.com>

On Mon, Mar 31, 2008 at 11:33 PM, Philip Brown <phil at bolthole.com> wrote:
> On Sun, Mar 30, 2008 at 11:16:32PM +0800, Aubrey Li wrote:
>  > Thanks Murray, I'll have a try.
>  > Just be curious, why don't we upgrade our system to solaris 10?
>
>
>  The mission of the CSW packaging effort can be started as providing
>  "supported packages of freeware", for "supported versions of solaris".
>
>  Solaris 8 is the oldest version of solaris fully supported by sun.
>  Therefore, all our packages have to support solaris 8 at this time.
>
>  This is mentioned with somewhat different wording, at
>  http://www.blastwave.org/standards/core_principles.html

That makes sense, I'll follow the principles.

Thanks for the info.
-Aubrey


From Joerg.Schilling at fokus.fraunhofer.de  Mon Mar 31 18:31:44 2008
From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling)
Date: Mon, 31 Mar 2008 18:31:44 +0200
Subject: [csw-dev] checkpkg problem
In-Reply-To: <6d6a94c50803310924n5c02cdffrcf882ef1cb90f1c1@mail.gmail.com>
References: <6d6a94c50803300446l75ee5120h8ef2db4af4e11f55@mail.gmail.com>
	<22901.1206883930@gerd>
	<6d6a94c50803300809v424b8c7bx7f1cf574aacf6252@mail.gmail.com>
	<6d6a94c50803302226m74bd433djf53f2bd44ad8d9c8@mail.gmail.com>
	<47f0f97c.70tVzGx/kXhfRnQB%Joerg.Schilling@fokus.fraunhofer.de>
	<6d6a94c50803310924n5c02cdffrcf882ef1cb90f1c1@mail.gmail.com>
Message-ID: <47f111f0.UPsS89Eupe9MgsYr%Joerg.Schilling@fokus.fraunhofer.de>

"Aubrey Li" <aubreylee at gmail.com> wrote:

> So do you mean the package has no problem, and we can ignore this
> message, can't we?
> If we can't ignore this message, how to fix this issue? I just used
> the script to create package.
> Is there any choice to make a TAR archive which makes Sun's cpio happy?
> or don't use Sun's cpio?

The SYSv package format is not expected to use TAR archives and it defaults to 
unpack things with Sun's cpio.

The real problem may be in Phil's unpack KSH script that tries to find the size 
of a header and then skips it. Without being able to look into the archive that 
cpio unpacks at that time, I cannot tell more.

Maybe this is Phil's task ;-)

J?rg

-- 
 EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin
       js at cs.tu-berlin.de                (uni)  
       schilling at fokus.fraunhofer.de     (work) Blog: http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily


From phil at bolthole.com  Mon Mar 31 19:20:43 2008
From: phil at bolthole.com (Philip Brown)
Date: Mon, 31 Mar 2008 10:20:43 -0700
Subject: [csw-dev] checkpkg problem
In-Reply-To: <47f111f0.UPsS89Eupe9MgsYr%Joerg.Schilling@fokus.fraunhofer.de>;
	from Joerg.Schilling@fokus.fraunhofer.de on Mon, Mar 31, 2008 at
	06:31:44PM +0200
References: <6d6a94c50803300446l75ee5120h8ef2db4af4e11f55@mail.gmail.com>
	<22901.1206883930@gerd>
	<6d6a94c50803300809v424b8c7bx7f1cf574aacf6252@mail.gmail.com>
	<6d6a94c50803302226m74bd433djf53f2bd44ad8d9c8@mail.gmail.com>
	<47f0f97c.70tVzGx/kXhfRnQB%Joerg.Schilling@fokus.fraunhofer.de>
	<6d6a94c50803310924n5c02cdffrcf882ef1cb90f1c1@mail.gmail.com>
	<47f111f0.UPsS89Eupe9MgsYr%Joerg.Schilling@fokus.fraunhofer.de>
Message-ID: <20080331102043.C27787@bolthole.com>

On Mon, Mar 31, 2008 at 06:31:44PM +0200, Joerg Schilling wrote:
> 
> The real problem may be in Phil's unpack KSH script that tries to find the size 
> of a header and then skips it. Without being able to look into the archive that 
> cpio unpacks at that time, I cannot tell more.
> 

yes, this is true.
The error from cpio is normal. Just ignore it.
After all, the next line is, "retrying".



From dclarke at blastwave.org  Mon Mar 24 17:47:42 2008
From: dclarke at blastwave.org (Dennis Clarke)
Date: Mon, 24 Mar 2008 12:47:42 -0400 (EDT)
Subject: [csw-dev] test the dev list !
Message-ID: <2283.72.39.133.97.1206377262.squirrel@mail.blastwave.org>





        THIS  PAGE  INTENTIONALLY  LEFT  BLANK
           yes .. a paradox if it were true


From jeff at cjsa.com  Mon Mar 24 18:13:02 2008
From: jeff at cjsa.com (Jeffery Small)
Date: Mon, 24 Mar 2008 17:13:02 GMT
Subject: [csw-dev] What's this list for?
Message-ID: <Jy8vtq.Iqx@cjsa.com>

Sorry, I must have missed some discussion along the way.  What is the dev list
supposed to be for as opposed to the maint list?  Thanks.

Regards,
-- 
Jeff

Jeffery Small  <jeff at blastwave.org>
In real life:  <jeff at cjsa.com>



From dam at blastwave.org  Mon Mar 24 21:48:48 2008
From: dam at blastwave.org (Dagobert Michelsen)
Date: Mon, 24 Mar 2008 21:48:48 +0100
Subject: [csw-dev] Multi-ISA GAR
In-Reply-To: <47E805E3.1060705@pleiades.fr.eu.org>
References: <20080319100038.C3150@bolthole.com>
	<33F3A8F3-6FA2-4E2E-91D2-6909A9902BCC@blastwave.org>
	<47E805E3.1060705@pleiades.fr.eu.org>
Message-ID: <F5645699-8B12-4DD0-BA05-FD383BE95C47@blastwave.org>

Hi Yann,

Am 24.03.2008 um 20:49 schrieb Yann Rouillard:
> I began to convert openssl to the new multi-ISA system, I have two
> comments so far:
>
> 	- the default isa install target is fine for standard programs  
> using automatic-multiarch-compile
> the classic configure script but will not work for everything (and
>
> indeed not for openssl).
> 	I added a new INSTALL_EXTRA_ISA_SCRIPTS to allow to redefine the  
> target
> used: see http://svn.blastwave.org/trac/changeset/1994

This is a good solution for openssl, but I don't believe
that there are any more packages built in the same way
(correct me if I'm wrong on this). I think it would be better
to solve this with

ifneq ($(ISA),$(ISA_DEFAULT))
INSTALL_SCRIPTS = $(INSTALL_EXTRA_ISA_SCRIPTS)
endif

inside the openssl-Makefile and not make a general rule out
of it. A more general solution would be

   INSTALL_SCRIPTS_$(ISA) ?= $(WORKSRC)/Makefile
   INSTALL_SCRIPTS ?= $(INSTALL_SCRIPTS_ISA)

with overwriting the specific install scripts on demand.
The thing that is disturbing me is the special handling of the default
ISA. In fact there are two default-ISAs: one for 32 bit and one for  
64 bit.
This makes defining the extra-ISAS a bit ugly IMHO.

> 	- I would like to be able to not use the automatic prototype  
> generation
> but I didn't see a way to disable your new automatic prototype
> generation. Could you add this option ?

This should work out-of-the-box. If the prototype is defined
in DISTFILES it shouldn't get overridden, so existing Makefiles
shouldn't break. Generally dynamic prototypes don't need adjustment
on version updates, so if there is anything missing which you need to
use them let me know what is missing and we may find a general solution.

> 	- I still have a problem building openssl, the first time I try to  
> build the package target after a "gmake clean", I have the  
> following error:
> 	gmake: *** No rule to make target `package-CSWossl', needed by  
> `package'.  Stop.
> 	If retry "gmake package" after this error, it works.
>
> 	You'll find the debug log of the first gmake call in my home:  
> ~yann/gmake_package.debug.log
> 	You'll find my automatic-multiarch-compile build tree on ra: ~yann/ 
> csw2/automatic-multiarch-compile/

This seems to be the same issue: dynamic prototypes. I'll make
a test environment with your openssl-Makefile and try to reproduce
this.


Best regards

   -- Dago


From dam at blastwave.org  Mon Mar 24 21:52:29 2008
From: dam at blastwave.org (Dagobert Michelsen)
Date: Mon, 24 Mar 2008 21:52:29 +0100
Subject: [csw-dev] What's this list for?
In-Reply-To: <Jy8vtq.Iqx@cjsa.com>
References: <Jy8vtq.Iqx@cjsa.com>
Message-ID: <5FC3DC1E-D19D-44F0-9E7E-1257BABC9A89@blastwave.org>

Hi Jeff,

Am 24.03.2008 um 17:13 schrieb Jeffery Small:
> Sorry, I must have missed some discussion along the way.  What is  
> the dev list
> supposed to be for as opposed to the maint list?  Thanks.

This list is for everything concerning package builds,
from compiler-flags to GAR, from pkg-get to Trac.
The only thing which should remain on the maintainer
list are discussions about people joining and leaving
and other things which may be better discussed in a
closed group.

The idea is that this dev-list is open for everybody
with an open mailing list archive while the maintainer
list stays closed to active maintainers.


Best regards

   -- Dago


From dam at blastwave.org  Mon Mar 24 22:20:05 2008
From: dam at blastwave.org (Dagobert Michelsen)
Date: Mon, 24 Mar 2008 22:20:05 +0100
Subject: [csw-dev] [libart 0002824]: RPATH contains
	/opt/csw/lib/sparcv9/$ISALIST
In-Reply-To: <20080324.20525100.1688352647@gyor.oxdrove.co.uk>
References: <200803211925.m2LJP8YB003508@callisto.blastwave.org>
	<20080321.19493100.2182619288@gyor.oxdrove.co.uk>
	<6C4B6E12-C46B-41F9-B1E3-7BFBFD8D8A9C@blastwave.org>
	<20080321.20314500.1513284418@gyor.oxdrove.co.uk>
	<16E0CBC5-1471-4C5C-818C-E27561981246@blastwave.org>
	<20080321.20470900.4124343341@gyor.oxdrove.co.uk>
	<836BD06B-8EF4-4148-87D8-51C6E10F070C@blastwave.org>
	<20080321.21201000.1964856037@gyor.oxdrove.co.uk>
	<C338E757-CD81-4093-AE1C-44FDA6BAC599@blastwave.org>
	<20080322.11171500.2414617030@gyor.oxdrove.co.uk>
	<75F0377E-79EE-47D6-983A-B2776010056E@blastwave.org>
	<20080324.20525100.1688352647@gyor.oxdrove.co.uk>
Message-ID: <7A398180-54F8-4E7F-B524-BDE40EA56EF4@blastwave.org>

Hi James.

I moved this to the new dev-list, as it may be worth
conserving the discussion :-) That's why I am keeping
most of the discussion fully quoted.

Am 24.03.2008 um 20:52 schrieb James Lee:
> On 24/03/08, 19:48:09, Dagobert Michelsen <dam at blastwave.org> wrote
>>>   export OPT="-fast -xnolibmopt -xspace -xtarget=ultra2 -xarch=v8
>>> -xstrconst"
>>>   make -e clean all
>>>   mv cswstage cswstage.generic
>>>
>>>   export OPT="-fast -xtarget=ultra3i -xarch=v8plusb -xstrconst"
>>>   export MLIB="-lm"
>>>   make -e clean all
>>>   mv cswstage cswstage.sparcv9+vis2
>
>> Ok, so you have different optimizer flags for each ISA, right?
>> And you have different exported variables per ISA.
>
> Yes.  The flags are not just about matching the instruction set.
> The flags are to suit the needs of the binaries.  Eg, a rarely used
> utility or one used for < 100ms is best optimised for size and
> certainly doesn't need arch variants.  The binaries that run for
> hours get everything I can give them.  It's not even true that the
> expected flags work best.
>
> I'm even putting v8 files in the v9 isaexec directory because
> Solaris 8 lacks the v8plusb isaexec entry (works with bins, not libs).

Looks quite like a hack... Any idea why the libs don't work?
I always thought the linker inspected the libs and decided
if the memory model is suitable instead of looking at the
directory name.

>> Why do you
>> use libm only for certain ISAs?
>
> Size optimisation.  It's really the inline libs I'm controlling
> here.

I see. So it would make sense to you to control the exported
variables per ISA?

>>> make actual calls a makefile that calls a special rmakefile, I  
>>> digress
>>> but somewhere in GAR it said make had to be gmake - whoa!!!
>
>> I'm sorry, I can't follow you here. Is gmake a problem?
>
> Yes when a makefile is written for make.  Now I could rewrite the
> makefile or I could use old make - easy choice.

If it is needed we can provide rules for ccsmake and xpg4make
which can easily selected if that helps you.

> Are yes, do we
> have imake capabilities?

Yes, those were inherited:

> # configure a package that uses imake
> # FIXME: untested and likely not the right way to handle the
> # argumentsconfigure-%/Imakefile:
>         @echo " ==> Running xmkmf in $*"
>         @cd $* && $(CONFIGURE_ENV) xmkmf $(CONFIGURE_ARGS)
>         @$(MAKECOOKIE)



> Maintainers with packages that just
> "configure && gmake install" are very lucky.

You can easily change behaviour to suit your needs with
   CONFIGURE_SCRIPTS = custom
   BUILD_SCRIPTS = custom

configure-custom: ...

build-custom: ...

>>> ARCHES="sparcv8 sparcv8plus+vis sparcv9+vis2"
>>> PROGS="rpict rvu rtrace ranimove ximage pfilt"
>>>
>>> for arch in $ARCHES
>>> do
>>>   mkdir -p `pwd`/cswstage/opt/csw/bin/$arch
>>>   for prog in $PROGS
>>>   do
>>>     # repeat in arch loop for simplicity...
>>>     rm -f cswstage/opt/csw/bin/$prog
>>>     cp cswstage.$arch/opt/csw/bin/$prog cswstage/opt/csw/bin/$arch/
>>> $prog
>>>   done
>>> done
>
>> Does PROGS contain bin/* or is it a subset?
>
> Just those 6, of about 125.  I definitely don't want 4 arch examples
> all.

Okay.

>> Additionally, you configure the binaries all to standard
>> locations (like bin/) and then move them to the ISA
>> subdirs. In the current multi-ISA GAR the individual
>> ISA-builds are configured to already use the ISA-subdirs
>> (like bindir=bin/sparcv9) which looks cleaner to me.
>
> That's where the current makefile puts them, it's not worth fighting,
> just go with the flow and move after.  There is no point it changing
> the build (and passing back) because only a packager wants the options
> and isaexec, anyone building for themselves knows their target.

Yes, but currently GAR resets bindir, libdir, etc. to the
ISA subdirectories automatically. This introduces some complexity
which I thought was necessary, but I am doubting now if this
could be reverted to the style you use. This would make the
GAR design quite a bit simpler.

> I know it takes the machine longer to build 5 variants and then
> only use a small number of each but the machine's time is cheap,
> mine is short (even if it's cheap ;-) ).
>
>>> ie, "manually", in this case it is almost regular but still a  
>>> country
>>> mile from wanting each example of each arch.  I really can't see an
>>> easier and clearer way of expressing these actions.
>
>> The steps are fairly straight-forward, and so a decent
>> abstraction could be added to GAR. I'll prepare something
>> for further discussion.
>
> Just keep in mind the need for flexibility.   I'm not against
> automation and build system in principle - the first thing I do with
> any project is waste 50% of the time working out how to do the job
> twice as quickly.  I am against systems that don't help.

I hereby allow you to kick me as long as you think
GAR doesn't help you ;-)

>>> For special optimisation, simplistically either speed or
>>> space, I want to be able use whichever binary benefits from each
>>> set of
>>> flags.
>
>> I see. Thanks for the input. Are you currently doing
>> these steps manually or do you have a script?
>
> I just record these steps in a script file, then run it.  It's all
> logged and repeatable and kept for the required 3 GPL years.
>
>
>
>> I would
>> like to convert this to GAR as a reference for my
>> "5 non-trivial packages" as it uses an extreme diversity
>> in compiler flags and ISA. As you have done quite some
>> work with compiler optiomization flags it would be great
>> if you could review the optimizer flag section in the
>> multi-ISA gar.conf.mk:
>>    <http://svn.blastwave.org/trac/browser/csw/branches/dam/automatic-
>> multiarch-compile/gar.conf.mk#L261>
>
>
> How about povray where I do profiling runs and feedback that back
> for recompilation?
>
> build(){
>
> for makefile in $(find . -name Makefile)
> do
>     gsed -i -e '/^CXXFLAGS.*/s//CXXFLAGS='"$FLAGS"' - 
> xprofile=collect/'
> $makefile
> done
>
>
> rm -rf povray.profile
> gmake clean
> gmake -j $(psrinfo | wc -l)
>
> for i in $(find scenes -name \*.pov)
> do
>     echo $i
>     echo ${i%/*}
>     unix/povray -D +A +L ./include/ +L ${i%/*}/ $i
> done
>
> for makefile in $(find . -name Makefile)
> do
>    gsed -i -e 's~^CXXFLAGS.*~CXXFLAGS='"$FLAGS"'
> -xprofile=use:'$(pwd)'\/povray~' $makefile
> done
>
> gmake clean
> gmake -j $(psrinfo | wc -l)
>
> mkdir -p cswstage/opt/csw/bin/${ARCH}
> cp unix/povray cswstage/opt/csw/bin/${ARCH}
>
> }
>
> FLAGS='-fast -xnolibmopt -xspace -xtarget=ss20 -xarch=v8'
> ARCH=sparcv8
> build
>
> FLAGS='-fast -xtarget=ultra2'
> ARCH=sparcv8plus+vis
> build
>
> if isalist | grep vis2 ; then
> FLAGS='-fast -xtarget=ultra3i -xarch=v8plusb'
> ARCH=sparcv9+vis2
> build
> fi

Fine. I'll give it a try when libiconv is finished and GAR
gets refined.


Thanks!

   -- Dago


From ihsan at blastwave.org  Mon Mar 24 22:36:50 2008
From: ihsan at blastwave.org (Ihsan Dogan)
Date: Mon, 24 Mar 2008 22:36:50 +0100
Subject: [csw-dev] What's this list for?
In-Reply-To: <Jy8vtq.Iqx@cjsa.com>
References: <Jy8vtq.Iqx@cjsa.com>
Message-ID: <47E81EF2.8040309@blastwave.org>

Am 24.3.2008 18:13 Uhr, Jeffery Small schrieb:
> Sorry, I must have missed some discussion along the way.  What is the dev list
> supposed to be for as opposed to the maint list?  Thanks.
>   
We were discussing that a few weeks ago on the maintainers list. Because 
I was abroad for two weeks, I didn't had any time to take care of this.

https://lists.blastwave.org/mailman/private/maintainers/2008-March/021245.html



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/



From yann at pleiades.fr.eu.org  Mon Mar 24 22:53:40 2008
From: yann at pleiades.fr.eu.org (Yann Rouillard)
Date: Mon, 24 Mar 2008 22:53:40 +0100
Subject: [csw-dev] Multi-ISA GAR
In-Reply-To: <F5645699-8B12-4DD0-BA05-FD383BE95C47@blastwave.org>
References: <20080319100038.C3150@bolthole.com>
	<33F3A8F3-6FA2-4E2E-91D2-6909A9902BCC@blastwave.org>
	<47E805E3.1060705@pleiades.fr.eu.org>
	<F5645699-8B12-4DD0-BA05-FD383BE95C47@blastwave.org>
Message-ID: <47E822E4.6080608@pleiades.fr.eu.org>

Dagobert Michelsen a ?crit :
> Hi Yann,
> 
> Am 24.03.2008 um 20:49 schrieb Yann Rouillard:
>> I began to convert openssl to the new multi-ISA system, I have two
>> comments so far:
>>
>>     - the default isa install target is fine for standard programs 
>> using automatic-multiarch-compile
>> the classic configure script but will not work for everything (and
>>
>> indeed not for openssl).
>>     I added a new INSTALL_EXTRA_ISA_SCRIPTS to allow to redefine the 
>> target
>> used: see http://svn.blastwave.org/trac/changeset/1994
> 
> This is a good solution for openssl, but I don't believe
> that there are any more packages built in the same way
> (correct me if I'm wrong on this). I think it would be better
> to solve this with

Well, I definitely think there are other packages like openssl, don't 
underestimate the imagination of developpers, althought it shouldn't be 
the majority.

In blastwave, I think at least pine and procmail don't use the standard 
configure, so it would be nicer to be able to handle that case without 
specific tweak.

> 
>>     - I would like to be able to not use the automatic prototype 
>> generation
>> but I didn't see a way to disable your new automatic prototype
>> generation. Could you add this option ?
> 
> This should work out-of-the-box. If the prototype is defined
> in DISTFILES it shouldn't get overridden, so existing Makefiles
> shouldn't break. Generally dynamic prototypes don't need adjustment
> on version updates, so if there is anything missing which you need to
> use them let me know what is missing and we may find a general solution.

The prototype are parts of DISTFILES, have a look at 
http://svn.blastwave.org/trac/browser/csw/branches/dam/automatic-multiarch-compile/lib/openssl/Makefile
They are arch suffixed, maybe that's not correctly handled.

>>     - I still have a problem building openssl, the first time I try to 
>> build the package target after a "gmake clean", I have the following 
>> error:
>>     gmake: *** No rule to make target `package-CSWossl', needed by 
>> `package'.  Stop.
>>     If retry "gmake package" after this error, it works.
>>
>>     You'll find the debug log of the first gmake call in my home: 
>> ~yann/gmake_package.debug.log
>>     You'll find my automatic-multiarch-compile build tree on ra: 
>> ~yann/csw2/automatic-multiarch-compile/
> 
> This seems to be the same issue: dynamic prototypes. I'll make
> a test environment with your openssl-Makefile and try to reproduce
> this.

Thanks.

Yann


From dam at blastwave.org  Mon Mar 24 23:15:52 2008
From: dam at blastwave.org (Dagobert Michelsen)
Date: Mon, 24 Mar 2008 23:15:52 +0100
Subject: [csw-dev] Multi-ISA GAR
In-Reply-To: <47E822E4.6080608@pleiades.fr.eu.org>
References: <20080319100038.C3150@bolthole.com>
	<33F3A8F3-6FA2-4E2E-91D2-6909A9902BCC@blastwave.org>
	<47E805E3.1060705@pleiades.fr.eu.org>
	<F5645699-8B12-4DD0-BA05-FD383BE95C47@blastwave.org>
	<47E822E4.6080608@pleiades.fr.eu.org>
Message-ID: <D85FED99-46EF-49A2-9B60-6C17067EB839@blastwave.org>

Hi Yann,

Am 24.03.2008 um 22:53 schrieb Yann Rouillard:
> Dagobert Michelsen a ?crit :
>> Am 24.03.2008 um 20:49 schrieb Yann Rouillard:
>>> I began to convert openssl to the new multi-ISA system, I have two
>>> comments so far:
>>>
>>>     - the default isa install target is fine for standard  
>>> programs using automatic-multiarch-compile
>>> the classic configure script but will not work for everything (and
>>>
>>> indeed not for openssl).
>>>     I added a new INSTALL_EXTRA_ISA_SCRIPTS to allow to redefine  
>>> the target
>>> used: see http://svn.blastwave.org/trac/changeset/1994
>>
>> This is a good solution for openssl, but I don't believe
>> that there are any more packages built in the same way
>> (correct me if I'm wrong on this). I think it would be better
>> to solve this with
>
> Well, I definitely think there are other packages like openssl,  
> don't underestimate the imagination of developpers, althought it  
> shouldn't be the majority.
>
> In blastwave, I think at least pine and procmail don't use the  
> standard configure, so it would be nicer to be able to handle that  
> case without specific tweak.

I am not against flexibility here. What I don't like
is the special meaning of ISA_DEFAULT here. You have
a custom installer, which you could call simply by
overriding the generic installer in your Makefile:

   install-isa-pentium_pro install-isa-amd64: install-isa
     $(DONADA)

If there really is need for the special case I'd like
this to be handled in a more general way working for
all ISAs, not only ISA_DEFAULT.


Best regards

   -- Dago

From dam at blastwave.org  Tue Mar 25 09:25:30 2008
From: dam at blastwave.org (Dagobert Michelsen)
Date: Tue, 25 Mar 2008 09:25:30 +0100
Subject: [csw-dev] Multi-ISA GAR
In-Reply-To: <47E822E4.6080608@pleiades.fr.eu.org>
References: <20080319100038.C3150@bolthole.com>
	<33F3A8F3-6FA2-4E2E-91D2-6909A9902BCC@blastwave.org>
	<47E805E3.1060705@pleiades.fr.eu.org>
	<F5645699-8B12-4DD0-BA05-FD383BE95C47@blastwave.org>
	<47E822E4.6080608@pleiades.fr.eu.org>
Message-ID: <87AF833D-9833-4B49-B832-043C9B94FFB2@blastwave.org>

Hi Yann,

Am 24.03.2008 um 22:53 schrieb Yann Rouillard:
>>>     I added a new INSTALL_EXTRA_ISA_SCRIPTS to allow to redefine  
>>> the target
>>> used: see http://svn.blastwave.org/trac/changeset/1994
>> This is a good solution for openssl, but I don't believe
>> that there are any more packages built in the same way
>> (correct me if I'm wrong on this). I think it would be better
>> to solve this with
>
> Well, I definitely think there are other packages like openssl,  
> don't underestimate the imagination of developpers, althought it  
> shouldn't be the majority.
>
> In blastwave, I think at least pine and procmail don't use the  
> standard configure, so it would be nicer to be able to handle that  
> case without specific tweak.

Now I see what you are going to accomplish: you
use the install-part of the Makefile _only_ for
the default ISA and copy in everything else manually.
That is somewhat different from what I had envisioned
but nonetheless useful. James again uses the install
differently: he has separate dirs for DESTDIR-installation
and copies the files together as last step. This may be
elegantly combinable with your approach. Please
continue while I give this some more thought :-)

Best regards

   -- Dago


From dam at blastwave.org  Tue Mar 25 22:00:44 2008
From: dam at blastwave.org (Dagobert Michelsen)
Date: Tue, 25 Mar 2008 22:00:44 +0100
Subject: [csw-dev] Multi-ISA GAR
In-Reply-To: <47E822E4.6080608@pleiades.fr.eu.org>
References: <20080319100038.C3150@bolthole.com>
	<33F3A8F3-6FA2-4E2E-91D2-6909A9902BCC@blastwave.org>
	<47E805E3.1060705@pleiades.fr.eu.org>
	<F5645699-8B12-4DD0-BA05-FD383BE95C47@blastwave.org>
	<47E822E4.6080608@pleiades.fr.eu.org>
Message-ID: <EA0A9D34-86ED-4417-A6AE-9C2D905AE2B3@blastwave.org>

Hi Yann,

Am 24.03.2008 um 22:53 schrieb Yann Rouillard:
> Dagobert Michelsen a ?crit :
>>>     I added a new INSTALL_EXTRA_ISA_SCRIPTS to allow to redefine  
>>> the target
>>> used: see http://svn.blastwave.org/trac/changeset/1994
>>>
>> This is a good solution for openssl, but I don't believe
>> that there are any more packages built in the same way
>> (correct me if I'm wrong on this). I think it would be better
>> to solve this with
>
> Well, I definitely think there are other packages like openssl,  
> don't underestimate the imagination of developpers, althought it  
> shouldn't be the majority.

After some more thinking I think this would be a good idea:
- configure for each ISA to the same standard locations
   (e. g. bindir without $ISA suffix)
- install each ISA to a different, ISA-specific tree
- user-settable scripts allow merging of these trees
   verbatim, select specific binaries or move them to
   $ISA subdirs

I'll may have some time next week to prepare something.

>>>     - I would like to be able to not use the automatic prototype  
>>> generation
>>> but I didn't see a way to disable your new automatic prototype
>>> generation. Could you add this option ?
>>
>> This should work out-of-the-box. If the prototype is defined
>> in DISTFILES it shouldn't get overridden, so existing Makefiles
>> shouldn't break. Generally dynamic prototypes don't need adjustment
>> on version updates, so if there is anything missing which you need to
>> use them let me know what is missing and we may find a general  
>> solution.
>
> The prototype are parts of DISTFILES, have a look at http:// 
> svn.blastwave.org/trac/browser/csw/branches/dam/automatic-multiarch- 
> compile/lib/openssl/Makefile
> They are arch suffixed, maybe that's not correctly handled.

The dependencies in gar.pkg.mk were too strict, so
your files got overwritten each time after install. This
should be fixed in r2023:
   <http://svn.blastwave.org/trac/changeset/2023>

>>>     - I still have a problem building openssl, the first time I  
>>> try to build the package target after a "gmake clean", I have the  
>>> following error:
>>>     gmake: *** No rule to make target `package-CSWossl', needed  
>>> by `package'.  Stop.
>>>     If retry "gmake package" after this error, it works.
>>>
>>>     You'll find the debug log of the first gmake call in my home:  
>>> ~yann/gmake_package.debug.log
>>>     You'll find my automatic-multiarch-compile build tree on ra:  
>>> ~yann/csw2/automatic-multiarch-compile/
>>
>> This seems to be the same issue: dynamic prototypes. I'll make
>> a test environment with your openssl-Makefile and try to reproduce
>> this.
>
> Thanks.

I couldn't verify this although I copied your tree. Please
check again with r2023. If the error is still there I see
what I can do on your tree directly.


Best regards

   -- Dago

From aubreylee at gmail.com  Sun Mar 30 11:44:33 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Sun, 30 Mar 2008 17:44:33 +0800
Subject: [csw-dev] gcc issue
Message-ID: <6d6a94c50803300244u69318b00v68aa5d4fe9d793f4@mail.gmail.com>

Hi list,

I saw this sentence in the build standard:
<Our standard compiler is the sun compiler(s),
which should be used in preference to gcc,
unless the software only compiles with gcc.>

Unfortunately, the package I want to maintain is only
compiled with gcc. But I couldn't found gcc on thor.blastwave.org.

Please tell me what I should do.

Many thanks,
-Aubrey


From william at wbonnet.net  Sun Mar 30 12:36:24 2008
From: william at wbonnet.net (William Bonnet)
Date: Sun, 30 Mar 2008 12:36:24 +0200
Subject: [csw-dev] gcc issue
In-Reply-To: <6d6a94c50803300244u69318b00v68aa5d4fe9d793f4@mail.gmail.com>
References: <6d6a94c50803300244u69318b00v68aa5d4fe9d793f4@mail.gmail.com>
Message-ID: <47EF6D28.7090204@wbonnet.net>

Hi
> Unfortunately, the package I want to maintain is only
> compiled with gcc. But I couldn't found gcc on thor.blastwave.org.
>
> Please tell me what I should do.
>   
gcc is in /opt/csw/gcc3/bin

cheers

-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.blastwave.org An OpenSolaris Community Site
http://www.guses.org     French speaking Solaris User Group




From aubreylee at gmail.com  Sun Mar 30 12:50:46 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Sun, 30 Mar 2008 18:50:46 +0800
Subject: [csw-dev] gcc issue
In-Reply-To: <47EF6D28.7090204@wbonnet.net>
References: <6d6a94c50803300244u69318b00v68aa5d4fe9d793f4@mail.gmail.com>
	<47EF6D28.7090204@wbonnet.net>
Message-ID: <6d6a94c50803300350h21d373a9p467e45c621189967@mail.gmail.com>

Hi William,

On Sun, Mar 30, 2008 at 6:36 PM, William Bonnet <william at wbonnet.net> wrote:
> Hi
>
> > Unfortunately, the package I want to maintain is only
>  > compiled with gcc. But I couldn't found gcc on thor.blastwave.org.
>  >
>  > Please tell me what I should do.
>  >
>  gcc is in /opt/csw/gcc3/bin
>
Thanks, I thought gcc is in /usr/sfw/bin, I got it now, ;-)

-Aubrey


From aubreylee at gmail.com  Sun Mar 30 13:29:30 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Sun, 30 Mar 2008 19:29:30 +0800
Subject: [csw-dev] getopt.h missing on solaris8
Message-ID: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>

I found thor.blastwave.org is a solaris8 system,

bash-2.03$ cat /etc/release
                       Solaris 8 2/02 s28x_u7wos_08a INTEL
           Copyright 2002 Sun Microsystems, Inc.  All Rights Reserved.
                           Assembled 18 December 2001

And hope I didn't miss it again, getopt.h and its lib is missing on this box.
what is the right way to fix this issue? it's already there on my nevada system.

Thanks,
-Aubrey


From aubreylee at gmail.com  Sun Mar 30 13:46:52 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Sun, 30 Mar 2008 19:46:52 +0800
Subject: [csw-dev] checkpkg problem
Message-ID: <6d6a94c50803300446l75ee5120h8ef2db4af4e11f55@mail.gmail.com>

I got the following error on my side.

Running checkpkg on pkgfile
Examining pmtools-20071116-SunOS5.11-i386-CSW.pkg
Looking for bad strings...
Extracting files for more detailed inspection...
CSWpmtools/pkginfo
63+0 records in
63+0 records out
2 blocks
ERROR: filename should start with -20071116-
20071116 pmtools-20071116-SunOS5.11-i386-CSW.pkg
ERROR: failed check.

here is my pkginfo
========
PKG=CSWpmtools
NAME=pmtools
VERSION=20071116
CATEGORY=application
VENDOR=http://www.lesswatts.org/patches/linux_acpi/ packaged for CSW
by Aubrey Li
HOTLINE=http://www.blastwave.org/bugtrack/
========

The problem is the following line in /opt/csw/bin/checkpkg
====
software=`sed -n 's/^NAME=\([^ -]*\) -.*$/\1/p' $TMPFILE`
====
software is assigned NULL by sed.

I don't think my pkginfo file has problem, so I double checked on the shell:
------------------------
[aubrey-sz330p at pmtools]sed -n 's/^VERSION=//p' pkginfo
20071116
------------------------
VERSION is no problem.

-----------------------
#sed -n 's/^NAME=\([^ -]*\) -.*$/\1/p' pkginfo
-----------------------
NAME return nothing.

Is this a problem in checkpkg? or I did something wrong?

I'm really appreciated any hints!

Thanks,
-Aubrey


From Murray.Jensen at csiro.au  Sun Mar 30 15:29:15 2008
From: Murray.Jensen at csiro.au (Murray.Jensen at csiro.au)
Date: Mon, 31 Mar 2008 00:29:15 +1100
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: Your message of "Sun, 30 Mar 2008 19:29:30 +0800"
	<6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com> 
Message-ID: <22844.1206883755@gerd>

On Sun, 30 Mar 2008 19:29:30 +0800, "Aubrey Li" writes:
>getopt.h and its lib is missing on this box.
>what is the right way to fix this issue?

Solaris 8 doesn't have getopt.h - or the gnu getopt* functions in libc.

Grab the source from gnulib and link it in with your application ...

	http://www.gnu.org/software/gnulib/

or direct to getopt module ...

	http://www.gnu.org/software/gnulib/MODULES.html#module=getopt

There are links for each of the following files

	getopt.in.h
	getopt.c
	getopt1.c
	getopt_int.h

but the first one is bogus - it should be getopt.h ... click on the
link and when you get the error page, go into the address bar and
edit the url such that getopt.in.h is changed to getopt.h.

I think those four files will be enough ... you will have to figure
out how to link them in - usually edit a Makefile, Makefile.in or
Makefile.am file.

This has been discussed in the maintainers list before - but its not
your fault you didnt find it - the archives arent easily searchable
(unless someone else has a searchable interface?). Cheers!
								Murray...
--
Murray Jensen, CSIRO Materials Science and Engineering Phone: +61 3 9545 2075
Private Bag 33, Clayton South 3169, Australia            Fax: +61 3 9544 1128
Internet: Murray.Jensen at csiro.au

To the extent permitted by law, CSIRO does not represent, warrant and/or
guarantee that the integrity of this communication has been maintained or
that the communication is free of errors, virus, interception or interference.

The information contained in this e-mail may be confidential or privileged.
Any unauthorised use or disclosure is prohibited. If you have received this
e-mail in error, please delete it immediately and notify Murray Jensen on
+61 3 9545 2075. Thank you.


From Murray.Jensen at csiro.au  Sun Mar 30 15:32:10 2008
From: Murray.Jensen at csiro.au (Murray.Jensen at csiro.au)
Date: Mon, 31 Mar 2008 00:32:10 +1100
Subject: [csw-dev] checkpkg problem
In-Reply-To: Your message of "Sun, 30 Mar 2008 19:46:52 +0800"
	<6d6a94c50803300446l75ee5120h8ef2db4af4e11f55@mail.gmail.com> 
Message-ID: <22901.1206883930@gerd>

On Sun, 30 Mar 2008 19:46:52 +0800, "Aubrey Li" writes:
>NAME=pmtools

NAME must have the "real" package name (i.e. not the system V CSW... name)
followed by " - " and a short description (the sed pattern should have
given it away). Cheers!
								Murray...
--
Murray Jensen, CSIRO Materials Science and Engineering Phone: +61 3 9545 2075
Private Bag 33, Clayton South 3169, Australia            Fax: +61 3 9544 1128
Internet: Murray.Jensen at csiro.au

To the extent permitted by law, CSIRO does not represent, warrant and/or
guarantee that the integrity of this communication has been maintained or
that the communication is free of errors, virus, interception or interference.

The information contained in this e-mail may be confidential or privileged.
Any unauthorised use or disclosure is prohibited. If you have received this
e-mail in error, please delete it immediately and notify Murray Jensen on
+61 3 9545 2075. Thank you.


From aubreylee at gmail.com  Sun Mar 30 17:09:33 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Sun, 30 Mar 2008 23:09:33 +0800
Subject: [csw-dev] checkpkg problem
In-Reply-To: <22901.1206883930@gerd>
References: <6d6a94c50803300446l75ee5120h8ef2db4af4e11f55@mail.gmail.com>
	<22901.1206883930@gerd>
Message-ID: <6d6a94c50803300809v424b8c7bx7f1cf574aacf6252@mail.gmail.com>

On Sun, Mar 30, 2008 at 9:32 PM,  <Murray.Jensen at csiro.au> wrote:
> On Sun, 30 Mar 2008 19:46:52 +0800, "Aubrey Li" writes:
>  >NAME=pmtools
>
>  NAME must have the "real" package name (i.e. not the system V CSW... name)
>  followed by " - " and a short description (the sed pattern should have
>  given it away). Cheers!
>                                                                 Murray...
Many thanks, I got it.

-Aubrey


From aubreylee at gmail.com  Sun Mar 30 17:16:32 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Sun, 30 Mar 2008 23:16:32 +0800
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <22844.1206883755@gerd>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<22844.1206883755@gerd>
Message-ID: <6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>

On Sun, Mar 30, 2008 at 9:29 PM,  <Murray.Jensen at csiro.au> wrote:
> On Sun, 30 Mar 2008 19:29:30 +0800, "Aubrey Li" writes:
>  >getopt.h and its lib is missing on this box.
>  >what is the right way to fix this issue?
>
>  Solaris 8 doesn't have getopt.h - or the gnu getopt* functions in libc.
>
>  Grab the source from gnulib and link it in with your application ...
>
>         http://www.gnu.org/software/gnulib/
>
>  or direct to getopt module ...
>
>         http://www.gnu.org/software/gnulib/MODULES.html#module=getopt
>
>  There are links for each of the following files
>
>         getopt.in.h
>         getopt.c
>         getopt1.c
>         getopt_int.h
>
>  but the first one is bogus - it should be getopt.h ... click on the
>  link and when you get the error page, go into the address bar and
>  edit the url such that getopt.in.h is changed to getopt.h.
>
>  I think those four files will be enough ... you will have to figure
>  out how to link them in - usually edit a Makefile, Makefile.in or
>  Makefile.am file.
>
>  This has been discussed in the maintainers list before - but its not
>  your fault you didnt find it - the archives arent easily searchable
>  (unless someone else has a searchable interface?). Cheers!
>                                                                 Murray...

Thanks Murray, I'll have a try.
Just be curious, why don't we upgrade our system to solaris 10?

-Aubrey


From Joerg.Schilling at fokus.fraunhofer.de  Sun Mar 30 17:28:47 2008
From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling)
Date: Sun, 30 Mar 2008 17:28:47 +0200
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
Message-ID: <47efb1af.1k1w0/oKWDynZgCC%Joerg.Schilling@fokus.fraunhofer.de>

"Aubrey Li" <aubreylee at gmail.com> wrote:

> I found thor.blastwave.org is a solaris8 system,
>
> bash-2.03$ cat /etc/release
>                        Solaris 8 2/02 s28x_u7wos_08a INTEL
>            Copyright 2002 Sun Microsystems, Inc.  All Rights Reserved.
>                            Assembled 18 December 2001
>
> And hope I didn't miss it again, getopt.h and its lib is missing on this box.
> what is the right way to fix this issue? it's already there on my nevada system.

getopt.h does not officieally exist, it is a private file written by RMS.

J?rg

-- 
 EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin
       js at cs.tu-berlin.de                (uni)  
       schilling at fokus.fraunhofer.de     (work) Blog: http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily


From dclarke at blastwave.org  Sun Mar 30 17:35:25 2008
From: dclarke at blastwave.org (Dennis Clarke)
Date: Sun, 30 Mar 2008 11:35:25 -0400 (EDT)
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<22844.1206883755@gerd>
	<6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>
Message-ID: <2056.72.39.133.97.1206891325.squirrel@mail.blastwave.org>


> On Sun, Mar 30, 2008 at 9:29 PM,  <Murray.Jensen at csiro.au> wrote:
>> On Sun, 30 Mar 2008 19:29:30 +0800, "Aubrey Li" writes:
>>  >getopt.h and its lib is missing on this box.
>>  >what is the right way to fix this issue?
>>
>>  Solaris 8 doesn't have getopt.h - or the gnu getopt* functions in libc.
>>
>>  Grab the source from gnulib and link it in with your application ...
>>
>>         http://www.gnu.org/software/gnulib/
>>
>>  or direct to getopt module ...
>>
>>         http://www.gnu.org/software/gnulib/MODULES.html#module=getopt
>>
>>  There are links for each of the following files
>>
>>         getopt.in.h
>>         getopt.c
>>         getopt1.c
>>         getopt_int.h
>>
>>  but the first one is bogus - it should be getopt.h ... click on the
>>  link and when you get the error page, go into the address bar and
>>  edit the url such that getopt.in.h is changed to getopt.h.
>>
>>  I think those four files will be enough ... you will have to figure
>>  out how to link them in - usually edit a Makefile, Makefile.in or
>>  Makefile.am file.
>>
>>  This has been discussed in the maintainers list before - but its not
>>  your fault you didnt find it - the archives arent easily searchable
>>  (unless someone else has a searchable interface?). Cheers!
>>                                                                 Murray...
>
> Thanks Murray, I'll have a try.
> Just be curious, why don't we upgrade our system to solaris 10?

Solaris 8 is scheduled to be dropped entirely into the "time and material"
support class by Sun for Q12009.  That is March 2009.  I hope to have all of
the Solaris 10 based infrastructure up to speed before then.

Dennis



From dclarke at blastwave.org  Sun Mar 30 17:38:29 2008
From: dclarke at blastwave.org (Dennis Clarke)
Date: Sun, 30 Mar 2008 11:38:29 -0400 (EDT)
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <47efb1af.1k1w0/oKWDynZgCC%Joerg.Schilling@fokus.fraunhofer.de>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<47efb1af.1k1w0/oKWDynZgCC%Joerg.Schilling@fokus.fraunhofer.de>
Message-ID: <2059.72.39.133.97.1206891509.squirrel@mail.blastwave.org>


> "Aubrey Li" <aubreylee at gmail.com> wrote:
>
>> I found thor.blastwave.org is a solaris8 system,
>>
>> bash-2.03$ cat /etc/release
>>                        Solaris 8 2/02 s28x_u7wos_08a INTEL
>>            Copyright 2002 Sun Microsystems, Inc.  All Rights Reserved.
>>                            Assembled 18 December 2001
>>
>> And hope I didn't miss it again, getopt.h and its lib is missing on this
>> box.
>> what is the right way to fix this issue? it's already there on my nevada
>> system.
>
> getopt.h does not officieally exist, it is a private file written by RMS.
>
> J?rg

In the category "J?rg will know for sure"  :-)

What is the correct way to do things? Officially ?

Dennis



From Joerg.Schilling at fokus.fraunhofer.de  Sun Mar 30 17:45:20 2008
From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling)
Date: Sun, 30 Mar 2008 17:45:20 +0200
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<22844.1206883755@gerd>
	<6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>
Message-ID: <47efb590.GEPvZ3W62E/R68/e%Joerg.Schilling@fokus.fraunhofer.de>

"Aubrey Li" <aubreylee at gmail.com> wrote:

> >  This has been discussed in the maintainers list before - but its not
> >  your fault you didnt find it - the archives arent easily searchable
> >  (unless someone else has a searchable interface?). Cheers!
> >                                                                 Murray...
>
> Thanks Murray, I'll have a try.
> Just be curious, why don't we upgrade our system to solaris 10?

Sources that asume the existence of a GNU/RMS private include file cannot
be called portable. You should inform the authors.

J?rg

-- 
 EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin
       js at cs.tu-berlin.de                (uni)  
       schilling at fokus.fraunhofer.de     (work) Blog: http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily


From Joerg.Schilling at fokus.fraunhofer.de  Sun Mar 30 17:50:15 2008
From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling)
Date: Sun, 30 Mar 2008 17:50:15 +0200
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <2059.72.39.133.97.1206891509.squirrel@mail.blastwave.org>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<47efb1af.1k1w0/oKWDynZgCC%Joerg.Schilling@fokus.fraunhofer.de>
	<2059.72.39.133.97.1206891509.squirrel@mail.blastwave.org>
Message-ID: <47efb6b7.COsZ6N/5WPc3uyjH%Joerg.Schilling@fokus.fraunhofer.de>

"Dennis Clarke" <dclarke at blastwave.org> wrote:

> > getopt.h does not officieally exist, it is a private file written by RMS.
> >
> > J?rg
>
> In the category "J?rg will know for sure"  :-)
>
> What is the correct way to do things? Officially ?

The getopt() prototype is in unistd.h

Most GNU influenced software tests for getopt_long() but forgets
about the fact that getopt.h is even more non-standard.


http://www.opengroup.org/onlinepubs/009695399/functions/getopt.html

is the official standard.

J?rg

-- 
 EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin
       js at cs.tu-berlin.de                (uni)  
       schilling at fokus.fraunhofer.de     (work) Blog: http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily


From aubreylee at gmail.com  Sun Mar 30 17:57:23 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Sun, 30 Mar 2008 23:57:23 +0800
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <2056.72.39.133.97.1206891325.squirrel@mail.blastwave.org>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<22844.1206883755@gerd>
	<6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>
	<2056.72.39.133.97.1206891325.squirrel@mail.blastwave.org>
Message-ID: <6d6a94c50803300857o302c8769i3668606ae7c0cf2e@mail.gmail.com>

On Sun, Mar 30, 2008 at 11:35 PM, Dennis Clarke <dclarke at blastwave.org> wrote:
>
>
>  > On Sun, Mar 30, 2008 at 9:29 PM,  <Murray.Jensen at csiro.au> wrote:
>  >> On Sun, 30 Mar 2008 19:29:30 +0800, "Aubrey Li" writes:
>  >>  >getopt.h and its lib is missing on this box.
>  >>  >what is the right way to fix this issue?
>  >>
>  >>  Solaris 8 doesn't have getopt.h - or the gnu getopt* functions in libc.
>  >>
>  >>  Grab the source from gnulib and link it in with your application ...
>  >>
>  >>         http://www.gnu.org/software/gnulib/
>  >>
>  >>  or direct to getopt module ...
>  >>
>  >>         http://www.gnu.org/software/gnulib/MODULES.html#module=getopt
>  >>
>  >>  There are links for each of the following files
>  >>
>  >>         getopt.in.h
>  >>         getopt.c
>  >>         getopt1.c
>  >>         getopt_int.h
>  >>
>  >>  but the first one is bogus - it should be getopt.h ... click on the
>  >>  link and when you get the error page, go into the address bar and
>  >>  edit the url such that getopt.in.h is changed to getopt.h.
>  >>
>  >>  I think those four files will be enough ... you will have to figure
>  >>  out how to link them in - usually edit a Makefile, Makefile.in or
>  >>  Makefile.am file.
>  >>
>  >>  This has been discussed in the maintainers list before - but its not
>  >>  your fault you didnt find it - the archives arent easily searchable
>  >>  (unless someone else has a searchable interface?). Cheers!
>  >>                                                                 Murray...
>  >
>  > Thanks Murray, I'll have a try.
>  > Just be curious, why don't we upgrade our system to solaris 10?
>
>  Solaris 8 is scheduled to be dropped entirely into the "time and material"
>  support class by Sun for Q12009.  That is March 2009.  I hope to have all of
>  the Solaris 10 based infrastructure up to speed before then.
>
>  Dennis
>

Of course I can't wait then, my account will be invalid in one month, ;-)

-Aubrey


From dclarke at blastwave.org  Sun Mar 30 18:22:05 2008
From: dclarke at blastwave.org (Dennis Clarke)
Date: Sun, 30 Mar 2008 12:22:05 -0400 (EDT)
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <47efb6b7.COsZ6N/5WPc3uyjH%Joerg.Schilling@fokus.fraunhofer.de>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<47efb1af.1k1w0/oKWDynZgCC%Joerg.Schilling@fokus.fraunhofer.de>
	<2059.72.39.133.97.1206891509.squirrel@mail.blastwave.org>
	<47efb6b7.COsZ6N/5WPc3uyjH%Joerg.Schilling@fokus.fraunhofer.de>
Message-ID: <2246.72.39.133.97.1206894125.squirrel@mail.blastwave.org>


> "Dennis Clarke" <dclarke at blastwave.org> wrote:
>
>> > getopt.h does not officieally exist, it is a private file written by
>> RMS.
>> >
>> > J?rg
>>
>> In the category "J?rg will know for sure"  :-)
>>
>> What is the correct way to do things? Officially ?
>
> The getopt() prototype is in unistd.h
>
> Most GNU influenced software tests for getopt_long() but forgets
> about the fact that getopt.h is even more non-standard.
>
>
> http://www.opengroup.org/onlinepubs/009695399/functions/getopt.html
>
> is the official standard.
>

I knew that all the standards of value are at the OpenGroup but was not sure
of the link. So I sat back and waited and sure enough .. you knew :-)

So the issue is with the upstream coders that target Linux users and that is
becoming a defacto way of things these days.

Dennis



From dclarke at blastwave.org  Sun Mar 30 18:29:23 2008
From: dclarke at blastwave.org (Dennis Clarke)
Date: Sun, 30 Mar 2008 12:29:23 -0400 (EDT)
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <6d6a94c50803300857o302c8769i3668606ae7c0cf2e@mail.gmail.com>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<22844.1206883755@gerd>
	<6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>
	<2056.72.39.133.97.1206891325.squirrel@mail.blastwave.org>
	<6d6a94c50803300857o302c8769i3668606ae7c0cf2e@mail.gmail.com>
Message-ID: <2317.72.39.133.97.1206894563.squirrel@mail.blastwave.org>


> On Sun, Mar 30, 2008 at 11:35 PM, Dennis Clarke <dclarke at blastwave.org>
> wrote:
>>
>>
>>  > On Sun, Mar 30, 2008 at 9:29 PM,  <Murray.Jensen at csiro.au> wrote:
>>  >> On Sun, 30 Mar 2008 19:29:30 +0800, "Aubrey Li" writes:
>>  >>  >getopt.h and its lib is missing on this box.
>>  >>  >what is the right way to fix this issue?
>>  >>
>>  >>  Solaris 8 doesn't have getopt.h - or the gnu getopt* functions in
>> libc.
>>  >>
>>  >>  Grab the source from gnulib and link it in with your application ...
>>  >>
>>  >>         http://www.gnu.org/software/gnulib/
>>  >>
>>  >>  or direct to getopt module ...
>>  >>
>>  >>         http://www.gnu.org/software/gnulib/MODULES.html#module=getopt
>>  >>
>>  >>  There are links for each of the following files
>>  >>
>>  >>         getopt.in.h
>>  >>         getopt.c
>>  >>         getopt1.c
>>  >>         getopt_int.h
>>  >>
>>  >>  but the first one is bogus - it should be getopt.h ... click on the
>>  >>  link and when you get the error page, go into the address bar and
>>  >>  edit the url such that getopt.in.h is changed to getopt.h.
>>  >>
>>  >>  I think those four files will be enough ... you will have to figure
>>  >>  out how to link them in - usually edit a Makefile, Makefile.in or
>>  >>  Makefile.am file.
>>  >>
>>  >>  This has been discussed in the maintainers list before - but its not
>>  >>  your fault you didnt find it - the archives arent easily searchable
>>  >>  (unless someone else has a searchable interface?). Cheers!
>>  >>                                                                 Murray...
>>  >
>>  > Thanks Murray, I'll have a try.
>>  > Just be curious, why don't we upgrade our system to solaris 10?
>>
>>  Solaris 8 is scheduled to be dropped entirely into the "time and
>> material"
>>  support class by Sun for Q12009.  That is March 2009.  I hope to have all
>> of
>>  the Solaris 10 based infrastructure up to speed before then.
>>
>>  Dennis
>>
>
> Of course I can't wait then, my account will be invalid in one month, ;-)

Say what ?  What account ?

Dennis



From aubreylee at gmail.com  Sun Mar 30 19:41:01 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Mon, 31 Mar 2008 01:41:01 +0800
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <2317.72.39.133.97.1206894563.squirrel@mail.blastwave.org>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<22844.1206883755@gerd>
	<6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>
	<2056.72.39.133.97.1206891325.squirrel@mail.blastwave.org>
	<6d6a94c50803300857o302c8769i3668606ae7c0cf2e@mail.gmail.com>
	<2317.72.39.133.97.1206894563.squirrel@mail.blastwave.org>
Message-ID: <6d6a94c50803301041p1a0e25dave254093344c58323@mail.gmail.com>

On Mon, Mar 31, 2008 at 12:29 AM, Dennis Clarke <dclarke at blastwave.org> wrote:
>
>
>  > On Sun, Mar 30, 2008 at 11:35 PM, Dennis Clarke <dclarke at blastwave.org>
>  > wrote:
>  >>
>  >>
>  >>  > On Sun, Mar 30, 2008 at 9:29 PM,  <Murray.Jensen at csiro.au> wrote:
>  >>  >> On Sun, 30 Mar 2008 19:29:30 +0800, "Aubrey Li" writes:
>  >>  >>  >getopt.h and its lib is missing on this box.
>  >>  >>  >what is the right way to fix this issue?
>  >>  >>
>  >>  >>  Solaris 8 doesn't have getopt.h - or the gnu getopt* functions in
>  >> libc.
>  >>  >>
>  >>  >>  Grab the source from gnulib and link it in with your application ...
>  >>  >>
>  >>  >>         http://www.gnu.org/software/gnulib/
>  >>  >>
>  >>  >>  or direct to getopt module ...
>  >>  >>
>  >>  >>         http://www.gnu.org/software/gnulib/MODULES.html#module=getopt
>  >>  >>
>  >>  >>  There are links for each of the following files
>  >>  >>
>  >>  >>         getopt.in.h
>  >>  >>         getopt.c
>  >>  >>         getopt1.c
>  >>  >>         getopt_int.h
>  >>  >>
>  >>  >>  but the first one is bogus - it should be getopt.h ... click on the
>  >>  >>  link and when you get the error page, go into the address bar and
>  >>  >>  edit the url such that getopt.in.h is changed to getopt.h.
>  >>  >>
>  >>  >>  I think those four files will be enough ... you will have to figure
>  >>  >>  out how to link them in - usually edit a Makefile, Makefile.in or
>  >>  >>  Makefile.am file.
>  >>  >>
>  >>  >>  This has been discussed in the maintainers list before - but its not
>  >>  >>  your fault you didnt find it - the archives arent easily searchable
>  >>  >>  (unless someone else has a searchable interface?). Cheers!
>  >>  >>                                                                 Murray...
>  >>  >
>  >>  > Thanks Murray, I'll have a try.
>  >>  > Just be curious, why don't we upgrade our system to solaris 10?
>  >>
>  >>  Solaris 8 is scheduled to be dropped entirely into the "time and
>  >> material"
>  >>  support class by Sun for Q12009.  That is March 2009.  I hope to have all
>  >> of
>  >>  the Solaris 10 based infrastructure up to speed before then.
>  >>
>  >>  Dennis
>  >>
>  >
>  > Of course I can't wait then, my account will be invalid in one month, ;-)
>
>  Say what ?  What account ?
For new maintainer, I have 30 days to create my first package, or my
account will expire.

Anyway, I made a simple link getopt staff in my application, and my
first package
was born successfully, ;-)

Thanks,
-Aubrey


From aubreylee at gmail.com  Mon Mar 31 07:26:18 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Mon, 31 Mar 2008 13:26:18 +0800
Subject: [csw-dev] checkpkg problem
In-Reply-To: <6d6a94c50803300809v424b8c7bx7f1cf574aacf6252@mail.gmail.com>
References: <6d6a94c50803300446l75ee5120h8ef2db4af4e11f55@mail.gmail.com>
	<22901.1206883930@gerd>
	<6d6a94c50803300809v424b8c7bx7f1cf574aacf6252@mail.gmail.com>
Message-ID: <6d6a94c50803302226m74bd433djf53f2bd44ad8d9c8@mail.gmail.com>

On Sun, Mar 30, 2008 at 11:09 PM, Aubrey Li <aubreylee at gmail.com> wrote:
> On Sun, Mar 30, 2008 at 9:32 PM,  <Murray.Jensen at csiro.au> wrote:
>  > On Sun, 30 Mar 2008 19:46:52 +0800, "Aubrey Li" writes:
>  >  >NAME=pmtools
>  >
>  >  NAME must have the "real" package name (i.e. not the system V CSW... name)
>  >  followed by " - " and a short description (the sed pattern should have
>  >  given it away). Cheers!
>  >                                                                 Murray...

I got another check error - cpio error, see below please.
the package is ok except this error.
any hints?

Thanks,
-Aubrey
====================================
[aubrey-sz330p at pmtools]createpkg -b
/export/home/aubrey/pkg-home/build/pmtools-20071116/cswstage/opt/csw
i copyright
## Building pkgmap from package prototype file.
## Processing pkginfo file.
WARNING: parameter <PSTAMP> set to "aubrey-sz330p20080331131732"
WARNING: parameter <CLASSES> set to "none"
## Attempting to volumize 5 entries in pkgmap.
part  1 -- 108 blocks, 15 entries
## Packaging one part.
/var/spool/pkg/CSWpmtools/pkgmap
/var/spool/pkg/CSWpmtools/pkginfo
/var/spool/pkg/CSWpmtools/reloc/bin/acpidump
/var/spool/pkg/CSWpmtools/reloc/bin/acpitbl
/var/spool/pkg/CSWpmtools/reloc/bin/acpixtract
/var/spool/pkg/CSWpmtools/reloc/bin/madt
/var/spool/pkg/CSWpmtools/install/copyright
/var/spool/pkg/CSWpmtools/install/depend
## Validating control scripts.
## Packaging complete.
Transferring <CSWpmtools> package instance
Conversion complete. Cleaning up in /var/spool/pkg

Running checkpkg on pkgfile
Examining pmtools-20071116,REV=2008.03.30-SunOS5.11-i386-CSW.pkg
Looking for bad strings...
Extracting files for more detailed inspection...
CSWpmtools/pkginfo
2 blocks
Info for pkg CSWpmtools
software= 'pmtools'
arch= 'i386'
desc='ACPI utilities'
pkgname='CSWpmtools'
version='20071116,REV=2008.03.30'
maintainer name='Aubrey Li'
maint email='aubrey at blastwave.org'
hotline='http://www.blastwave.org/bugtrack/'
basedir='/opt/csw'

Extracing pkg for examination of files...
cpio: Impossible header type.         <============= here
1 errors
retrying extract with different archive offset...
pkginfo
pkgmap
install
install/copyright
install/depend
reloc
reloc/bin
reloc/bin/acpidump
reloc/bin/acpitbl
reloc/bin/acpixtract
reloc/bin/madt
89 blocks
Checking uninstalled directory format package <CSWpmtools> from
</tmp/dissect.102987>
## Checking control scripts.
## Checking package objects.
## Checking is complete.
/usr/sbin/pkgchk passed.
libraries used are:
libc.so.1
cross-referencing with depend file (May take a while)

Examining 'depend' file
system      CSWcommon common - common files and dirs for CSW packages
system      SUNWcsl Core Solaris, (Shared Libs)
system      SUNWdpl Developer Profiled Libraries
SUGGESTION: you may want to add some or all of the following as depends:
   (Feel free to ignore SUNW or SPRO packages)
> SUNWcslr

Checking relocation ability...
trivial check passed

packaging of files under
/export/home/aubrey/pkg-home/build/pmtools-20071116/cswstage/opt/csw
complete
gzipping pkgfile


From Joerg.Schilling at fokus.fraunhofer.de  Mon Mar 31 16:47:24 2008
From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling)
Date: Mon, 31 Mar 2008 16:47:24 +0200
Subject: [csw-dev] checkpkg problem
In-Reply-To: <6d6a94c50803302226m74bd433djf53f2bd44ad8d9c8@mail.gmail.com>
References: <6d6a94c50803300446l75ee5120h8ef2db4af4e11f55@mail.gmail.com>
	<22901.1206883930@gerd>
	<6d6a94c50803300809v424b8c7bx7f1cf574aacf6252@mail.gmail.com>
	<6d6a94c50803302226m74bd433djf53f2bd44ad8d9c8@mail.gmail.com>
Message-ID: <47f0f97c.70tVzGx/kXhfRnQB%Joerg.Schilling@fokus.fraunhofer.de>

"Aubrey Li" <aubreylee at gmail.com> wrote:

> On Sun, Mar 30, 2008 at 11:09 PM, Aubrey Li <aubreylee at gmail.com> wrote:
> > On Sun, Mar 30, 2008 at 9:32 PM,  <Murray.Jensen at csiro.au> wrote:
> >  > On Sun, 30 Mar 2008 19:46:52 +0800, "Aubrey Li" writes:
> >  >  >NAME=pmtools
> >  >
> >  >  NAME must have the "real" package name (i.e. not the system V CSW... name)
> >  >  followed by " - " and a short description (the sed pattern should have
> >  >  given it away). Cheers!
> >  >                                                                 Murray...
>
> I got another check error - cpio error, see below please.
> the package is ok except this error.
> any hints?
>
> desc='ACPI utilities'
> pkgname='CSWpmtools'
> version='20071116,REV=2008.03.30'
> maintainer name='Aubrey Li'
> maint email='aubrey at blastwave.org'
> hotline='http://www.blastwave.org/bugtrack/'
> basedir='/opt/csw'
>
> Extracing pkg for examination of files...
> cpio: Impossible header type.         <============= here

This message is usually seen when Sun's cpio tries to extract a POSIXly correct
TAR archive.

J?rg

-- 
 EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin
       js at cs.tu-berlin.de                (uni)  
       schilling at fokus.fraunhofer.de     (work) Blog: http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily


From phil at bolthole.com  Mon Mar 31 17:33:56 2008
From: phil at bolthole.com (Philip Brown)
Date: Mon, 31 Mar 2008 08:33:56 -0700
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>;
	from aubreylee@gmail.com on Sun, Mar 30, 2008 at 11:16:32PM +0800
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<22844.1206883755@gerd>
	<6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>
Message-ID: <20080331083356.A99037@bolthole.com>

On Sun, Mar 30, 2008 at 11:16:32PM +0800, Aubrey Li wrote:
> Thanks Murray, I'll have a try.
> Just be curious, why don't we upgrade our system to solaris 10?


The mission of the CSW packaging effort can be started as providing
"supported packages of freeware", for "supported versions of solaris".

Solaris 8 is the oldest version of solaris fully supported by sun.
Therefore, all our packages have to support solaris 8 at this time.

This is mentioned with somewhat different wording, at
http://www.blastwave.org/standards/core_principles.html


From aubreylee at gmail.com  Mon Mar 31 18:24:15 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Tue, 1 Apr 2008 00:24:15 +0800
Subject: [csw-dev] checkpkg problem
In-Reply-To: <47f0f97c.70tVzGx/kXhfRnQB%Joerg.Schilling@fokus.fraunhofer.de>
References: <6d6a94c50803300446l75ee5120h8ef2db4af4e11f55@mail.gmail.com>
	<22901.1206883930@gerd>
	<6d6a94c50803300809v424b8c7bx7f1cf574aacf6252@mail.gmail.com>
	<6d6a94c50803302226m74bd433djf53f2bd44ad8d9c8@mail.gmail.com>
	<47f0f97c.70tVzGx/kXhfRnQB%Joerg.Schilling@fokus.fraunhofer.de>
Message-ID: <6d6a94c50803310924n5c02cdffrcf882ef1cb90f1c1@mail.gmail.com>

On Mon, Mar 31, 2008 at 10:47 PM, Joerg Schilling
<Joerg.Schilling at fokus.fraunhofer.de> wrote:
> "Aubrey Li" <aubreylee at gmail.com> wrote:
>
>  > On Sun, Mar 30, 2008 at 11:09 PM, Aubrey Li <aubreylee at gmail.com> wrote:
>  > > On Sun, Mar 30, 2008 at 9:32 PM,  <Murray.Jensen at csiro.au> wrote:
>  > >  > On Sun, 30 Mar 2008 19:46:52 +0800, "Aubrey Li" writes:
>  > >  >  >NAME=pmtools
>  > >  >
>  > >  >  NAME must have the "real" package name (i.e. not the system V CSW... name)
>  > >  >  followed by " - " and a short description (the sed pattern should have
>  > >  >  given it away). Cheers!
>  > >  >                                                                 Murray...
>  >
>  > I got another check error - cpio error, see below please.
>  > the package is ok except this error.
>  > any hints?
>  >
>
> > desc='ACPI utilities'
>  > pkgname='CSWpmtools'
>  > version='20071116,REV=2008.03.30'
>  > maintainer name='Aubrey Li'
>  > maint email='aubrey at blastwave.org'
>  > hotline='http://www.blastwave.org/bugtrack/'
>  > basedir='/opt/csw'
>  >
>  > Extracing pkg for examination of files...
>  > cpio: Impossible header type.         <============= here
>
>  This message is usually seen when Sun's cpio tries to extract a POSIXly correct
>  TAR archive.
>
So do you mean the package has no problem, and we can ignore this
message, can't we?
If we can't ignore this message, how to fix this issue? I just used
the script to create package.
Is there any choice to make a TAR archive which makes Sun's cpio happy?
or don't use Sun's cpio?

Thanks,
-Aubrey


From aubreylee at gmail.com  Mon Mar 31 18:25:36 2008
From: aubreylee at gmail.com (Aubrey Li)
Date: Tue, 1 Apr 2008 00:25:36 +0800
Subject: [csw-dev] getopt.h missing on solaris8
In-Reply-To: <20080331083356.A99037@bolthole.com>
References: <6d6a94c50803300429j4f81143cu572a3baf526a7dbf@mail.gmail.com>
	<22844.1206883755@gerd>
	<6d6a94c50803300816n72583243n1cc3c3418c87d4bf@mail.gmail.com>
	<20080331083356.A99037@bolthole.com>
Message-ID: <6d6a94c50803310925y5fadef0ak7e54bf3b02f02fec@mail.gmail.com>

On Mon, Mar 31, 2008 at 11:33 PM, Philip Brown <phil at bolthole.com> wrote:
> On Sun, Mar 30, 2008 at 11:16:32PM +0800, Aubrey Li wrote:
>  > Thanks Murray, I'll have a try.
>  > Just be curious, why don't we upgrade our system to solaris 10?
>
>
>  The mission of the CSW packaging effort can be started as providing
>  "supported packages of freeware", for "supported versions of solaris".
>
>  Solaris 8 is the oldest version of solaris fully supported by sun.
>  Therefore, all our packages have to support solaris 8 at this time.
>
>  This is mentioned with somewhat different wording, at
>  http://www.blastwave.org/standards/core_principles.html

That makes sense, I'll follow the principles.

Thanks for the info.
-Aubrey


From Joerg.Schilling at fokus.fraunhofer.de  Mon Mar 31 18:31:44 2008
From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling)
Date: Mon, 31 Mar 2008 18:31:44 +0200
Subject: [csw-dev] checkpkg problem
In-Reply-To: <6d6a94c50803310924n5c02cdffrcf882ef1cb90f1c1@mail.gmail.com>
References: <6d6a94c50803300446l75ee5120h8ef2db4af4e11f55@mail.gmail.com>
	<22901.1206883930@gerd>
	<6d6a94c50803300809v424b8c7bx7f1cf574aacf6252@mail.gmail.com>
	<6d6a94c50803302226m74bd433djf53f2bd44ad8d9c8@mail.gmail.com>
	<47f0f97c.70tVzGx/kXhfRnQB%Joerg.Schilling@fokus.fraunhofer.de>
	<6d6a94c50803310924n5c02cdffrcf882ef1cb90f1c1@mail.gmail.com>
Message-ID: <47f111f0.UPsS89Eupe9MgsYr%Joerg.Schilling@fokus.fraunhofer.de>

"Aubrey Li" <aubreylee at gmail.com> wrote:

> So do you mean the package has no problem, and we can ignore this
> message, can't we?
> If we can't ignore this message, how to fix this issue? I just used
> the script to create package.
> Is there any choice to make a TAR archive which makes Sun's cpio happy?
> or don't use Sun's cpio?

The SYSv package format is not expected to use TAR archives and it defaults to 
unpack things with Sun's cpio.

The real problem may be in Phil's unpack KSH script that tries to find the size 
of a header and then skips it. Without being able to look into the archive that 
cpio unpacks at that time, I cannot tell more.

Maybe this is Phil's task ;-)

J?rg

-- 
 EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin
       js at cs.tu-berlin.de                (uni)  
       schilling at fokus.fraunhofer.de     (work) Blog: http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily


From phil at bolthole.com  Mon Mar 31 19:20:43 2008
From: phil at bolthole.com (Philip Brown)
Date: Mon, 31 Mar 2008 10:20:43 -0700
Subject: [csw-dev] checkpkg problem
In-Reply-To: <47f111f0.UPsS89Eupe9MgsYr%Joerg.Schilling@fokus.fraunhofer.de>;
	from Joerg.Schilling@fokus.fraunhofer.de on Mon, Mar 31, 2008 at
	06:31:44PM +0200
References: <6d6a94c50803300446l75ee5120h8ef2db4af4e11f55@mail.gmail.com>
	<22901.1206883930@gerd>
	<6d6a94c50803300809v424b8c7bx7f1cf574aacf6252@mail.gmail.com>
	<6d6a94c50803302226m74bd433djf53f2bd44ad8d9c8@mail.gmail.com>
	<47f0f97c.70tVzGx/kXhfRnQB%Joerg.Schilling@fokus.fraunhofer.de>
	<6d6a94c50803310924n5c02cdffrcf882ef1cb90f1c1@mail.gmail.com>
	<47f111f0.UPsS89Eupe9MgsYr%Joerg.Schilling@fokus.fraunhofer.de>
Message-ID: <20080331102043.C27787@bolthole.com>

On Mon, Mar 31, 2008 at 06:31:44PM +0200, Joerg Schilling wrote:
> 
> The real problem may be in Phil's unpack KSH script that tries to find the size 
> of a header and then skips it. Without being able to look into the archive that 
> cpio unpacks at that time, I cannot tell more.
> 

yes, this is true.
The error from cpio is normal. Just ignore it.
After all, the next line is, "retrying".