multilibbing libgfortran

multilibbing libgfortran

[Andreas Jaeger]

[-- Attachment #1: Type: text/plain, Size: 394 bytes --]


libgfortran on the tree-ssa branch is not multi-libbed.  Is there
anywhere a description what needs to be done to enable multi-libs?  Or
who can step up to multi-lib the directory?

Andreas
-- 
 Andreas Jaeger, aj@suse.de, http://www.suse.de/~aj
  SuSE Linux AG, Deutschherrnstr. 15-19, 90429 Nürnberg, Germany
   GPG fingerprint = 93A3 365E CE47 B889 DF7F  FED1 389A 563C C272 A126

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]

Re: multilibbing libgfortran

[Alexandre Oliva]

On Aug  3, 2003, Andreas Jaeger <aj@suse.de> wrote:

> libgfortran on the tree-ssa branch is not multi-libbed.  Is there
> anywhere a description what needs to be done to enable multi-libs?

Ideally, nothing.  Nathanael said he was looking into moving
multilibbing of target libraries to the top level.  Perhaps we could
live on without a multilibbed libgfortran until this change is
implemented?

-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                 aoliva@{redhat.com, gcc.gnu.org}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist                Professional serial bug killer

Re: multilibbing libgfortran

[Andreas Jaeger]

[-- Attachment #1: Type: text/plain, Size: 765 bytes --]

Alexandre Oliva <aoliva@redhat.com> writes:

> On Aug  3, 2003, Andreas Jaeger <aj@suse.de> wrote:
>
>> libgfortran on the tree-ssa branch is not multi-libbed.  Is there
>> anywhere a description what needs to be done to enable multi-libs?
>
> Ideally, nothing.  Nathanael said he was looking into moving
> multilibbing of target libraries to the top level.  Perhaps we could
> live on without a multilibbed libgfortran until this change is
> implemented?

Depends how long it will take Nathanael to do it.  What's the best
estimate for this?

Andreas
-- 
 Andreas Jaeger, aj@suse.de, http://www.suse.de/~aj
  SuSE Linux AG, Deutschherrnstr. 15-19, 90429 Nürnberg, Germany
   GPG fingerprint = 93A3 365E CE47 B889 DF7F  FED1 389A 563C C272 A126

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]

Re: multilibbing libgfortran

[Phil Edwards]

On Sun, Aug 03, 2003 at 04:52:15PM +0200, Andreas Jaeger wrote:
> 
> So, what needs to be done to get this multi-libbed?  I tried copying
> some stuff from libstdc++-v3 and failed with automake problems :-(

Let me finish the libstdc++-v3 autotools conversion, and then it might
(or might not) be easier.

Phil

-- 
If ye love wealth greater than liberty, the tranquility of servitude greater
than the animating contest for freedom, go home and leave us in peace.  We seek
not your counsel, nor your arms.  Crouch down and lick the hand that feeds you;
and may posterity forget that ye were our countrymen.            - Samuel Adams

Re: multilibbing libgfortran

[Andreas Jaeger]

[-- Attachment #1: Type: text/plain, Size: 1070 bytes --]

Nathanael Nerode <neroden@twcny.rr.com> writes:

> Andreas Jaeger wrote:
>> Nathanael Nerode <neroden@twcny.rr.com> writes:
>> Unfortunately exact enough - but too late.  I'd prefer to have
>> libgfortran earlier multi-libbed if possible so that it can get better
>> tested.
>> So, what needs to be done to get this multi-libbed?  I tried copying
>> some stuff from libstdc++-v3 and failed with automake problems :-(
>> Andreas
>
> The current multilibbing scheme is demented, in my opinion.  I still
> recommend not multilibbing at the moment; when multilibbing is moved
> to the top level, you'll have to rip all the multilib code back *out*
> of gfortran.  You can still test gfortran with each individual library

I see.

> configuration directly, you know, right?

No, I don't know how I can build on x86_64 a 32-bit multilib directly.

Andreas
-- 
 Andreas Jaeger, aj@suse.de, http://www.suse.de/~aj
  SuSE Linux AG, Deutschherrnstr. 15-19, 90429 Nürnberg, Germany
   GPG fingerprint = 93A3 365E CE47 B889 DF7F  FED1 389A 563C C272 A126

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]

Re: multilibbing libgfortran

[Nathanael Nerode]

Andreas Jaeger wrote:
> Nathanael Nerode <neroden@twcny.rr.com> writes:
> 
> 
> Unfortunately exact enough - but too late.  I'd prefer to have
> libgfortran earlier multi-libbed if possible so that it can get better
> tested.
> 
> So, what needs to be done to get this multi-libbed?  I tried copying
> some stuff from libstdc++-v3 and failed with automake problems :-(
> 
> Andreas

The current multilibbing scheme is demented, in my opinion.  I still 
recommend not multilibbing at the moment; when multilibbing is moved to 
the top level, you'll have to rip all the multilib code back *out* of 
gfortran.  You can still test gfortran with each individual library 
configuration directly, you know, right?

Re: multilibbing libgfortran

[Andreas Jaeger]

[-- Attachment #1: Type: text/plain, Size: 1442 bytes --]

Nathanael Nerode <neroden@twcny.rr.com> writes:

>> Alexandre Oliva <aoliva@redhat.com> writes:
>>
>>> On Aug  3, 2003, Andreas Jaeger <aj@suse.de> wrote:
>>>
>>>> libgfortran on the tree-ssa branch is not multi-libbed.  Is there
>>>> anywhere a description what needs to be done to enable multi-libs?
>>>
>>> Ideally, nothing.  Nathanael said he was looking into moving
>>> multilibbing of target libraries to the top level.  Perhaps we could
>>> live on without a multilibbed libgfortran until this change is
>>> implemented?
>> Depends how long it will take Nathanael to do it.  What's the best
>> estimate for this?
>
> It should be in for 3.5.  I think it's probably too big a change for
> 3.4.  At the moment I don't anticipate it getting done before stage 1
> of 3.5 starts, as I still have other stuff I'm trying to do in stage 2
> of 3.4, and this is going to be a busy month for me.  That a good
> enough guess? :-/

Unfortunately exact enough - but too late.  I'd prefer to have
libgfortran earlier multi-libbed if possible so that it can get better
tested.

So, what needs to be done to get this multi-libbed?  I tried copying
some stuff from libstdc++-v3 and failed with automake problems :-(

Andreas
-- 
 Andreas Jaeger, aj@suse.de, http://www.suse.de/~aj
  SuSE Linux AG, Deutschherrnstr. 15-19, 90429 Nürnberg, Germany
   GPG fingerprint = 93A3 365E CE47 B889 DF7F  FED1 389A 563C C272 A126

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]

Re: multilibbing libgfortran

[Nathanael Nerode]

> Alexandre Oliva <aoliva@redhat.com> writes:
> 
>> On Aug  3, 2003, Andreas Jaeger <aj@suse.de> wrote:
>>
>>> libgfortran on the tree-ssa branch is not multi-libbed.  Is there
>>> anywhere a description what needs to be done to enable multi-libs?
>>
>> Ideally, nothing.  Nathanael said he was looking into moving
>> multilibbing of target libraries to the top level.  Perhaps we could
>> live on without a multilibbed libgfortran until this change is
>> implemented?
> 
> Depends how long it will take Nathanael to do it.  What's the best
> estimate for this?

It should be in for 3.5.  I think it's probably too big a change for 
3.4.  At the moment I don't anticipate it getting done before stage 1 of 
3.5 starts, as I still have other stuff I'm trying to do in stage 2 of 
3.4, and this is going to be a busy month for me.  That a good enough 
guess? :-/