help-octave
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: compiling SuiteSparse


From: Sergei Steshenko
Subject: Re: compiling SuiteSparse
Date: Sun, 27 May 2012 01:24:21 -0700 (PDT)




----- Original Message -----
> From: marco atzeri <address@hidden>
> To: "address@hidden" <address@hidden>
> Cc: 
> Sent: Sunday, May 27, 2012 8:07 AM
> Subject: Re: compiling SuiteSparse
> 
> On 5/27/2012 3:46 AM, Sergei Steshenko wrote:
>>  Hello,
>> 
>>  compiling SuiteSparse happens to be tricky, and I think I've done it 
> successfully, so I want to share my findings.
>> 
>>  These findings apply to both SuiteSparse-3.7.0, SuiteSparse-3.7.1; Octave 
> 'make check' completes successfully (if it matters). Also, TBB is used 
> during compilation and linking.
>> 
> 
> you forgot to mention which system, compiler and lapack, atlas
> versions you are using.
> Just to avoid confusion for system with different systems.
> 
> As Metis is not FREE for distribution, we usually do not
> suggest linking against it.
> 
>>  Regards,
>>     Sergei.
>> 
> 
> Regards
> Marco

Versions of all the dependencies, including the compiler, can be seen in the 
attached to the original Email 'UFconfig.mk' file.

The OS is 32 bit OpenSuse-11.1, but it really doesn't matter since I'm building 
practically everything from sources.

And I am especially glad and proud that successful compilation and linking 
against 'metis' is yet another victory of me and hopefully other people and 
pragmatism over the stupid GPL propaganda. And I've been building with 'metis' 
for years.

I remind that copying a binary inside an organization is _not_ distribution 
according to GPL FAQ.

If anybody is interested, I can send a snapshot of my AppsGromScratch which 
builds octave-3.6.1 and 77 packages - the latter can be installed globally 
and/or locally. Since building will be done in the person's computer, there 
will be no distribution at all.

Regards,
  Sergei.



reply via email to

[Prev in Thread] Current Thread [Next in Thread]