[Fwd: Re: Someone broke complex arithmetic]

[Emil Block]

-------- Original Message --------
Subject: 	Re: Someone broke complex arithmetic
Date: 	Mon, 25 Jul 2005 19:17:30 -0400
From: 	Emil Block <blime@cox.net>
To: 	Steve Kargl <sgk@troutmask.apl.washington.edu>
CC: 	fortran@gcc.gnu.org, gcc-bugzilla@gcc.gnu.org
References: 	<42DA8C72.2020207@wanadoo.fr> 
<200507172119.j6HLJPd26676@makai.watson.ibm.com> 
<42DAD899.2090403@wanadoo.fr> 
<200507172332.j6HNWmd27626@makai.watson.ibm.com> 
<42DB3313.2070707@wanadoo.fr> 
<200507182101.j6IL1fd27934@makai.watson.ibm.com> 
<42DC9232.10807@wanadoo.fr> <42DCADC4.9030405@gmail.com> 
<Pine.BSF.4.62.0507251248380.21428@pulcherrima.dbai.tuwien.ac.at> 
<1122294714.19360.131.camel@george.gkn-whl.co.uk> 
<20050725161527.GA35898@troutmask.apl.washington.edu>



Steve

For what its worth on 2 subjects :

1 -- When first compiling gfortran -- encountered problem with MPFR 
related to 32 bit versus 64 bit, and finally resolved it with switch 
settings for the build of MPFR, and did not report a bug, but it was 
akward.

2 -- Unformatted record header. As an interim solution while awaiting a 
fix for Bug 19303,  simply replaced "gfc_offset" with "long" at every 
occurrence in /gcc/libgfortan/io/transfer.c.  This provides 
compatibility across the board with gcc, c++, gfortran, and g77 
unformatted files
 (gcc 3.3.5 and 4.0 and 4.1).

blime

Steve Kargl wrote:

>On Mon, Jul 25, 2005 at 01:31:54PM +0100, Steve Edwards wrote:
>  
>
>>On Mon, 2005-07-25 at 11:49, Gerald Pfeifer wrote:
>>    
>>
>>>On Tue, 19 Jul 2005, FX Coudert wrote:
>>>      
>>>
>>>>>Don't folk run the gfortran testsuite???
>>>>>          
>>>>>
>>>>No. People don't regtest with gfortran enabled. That's a pity, since it only
>>>>adds little time to the total build and testing time.
>>>>        
>>>>
>>>I believe on of the reasons people often do not build with gfortran 
>>>enabled is that this requires additional libraries which are not part
>>>of common system configurations and do not come with GCC either.
>>>      
>>>
>>Having spent 46 hours bootstrapping gcc 4.0.1 on an elderly
>>rs6000-ibm-aix4.3.0.0 only to get a build failure with the mpfr library
>>(again...) I do tend to sympathise with this.
>>
>>    
>>
>
>Odd. If this is a recurring problem, one would think that someone
>would have filed a bug report.  A search of bugzilla with mpfr
>and MPFR shows no problems are reported.
>
>  
>