Re: [ublas] g++ 4.4.6 and boost ublas 1_48 compile problem

Subject: Re: [ublas] g++ 4.4.6 and boost ublas 1_48 compile problem
From: Umut Tabak (u.tabak_at_[hidden])
Date: 2012-02-22 11:33:41

On 02/22/2012 05:21 PM, Thomas Klimpel wrote:
>
> Oh, I don't think that the version of boost has anything to do with this. The problem should be the same in boost_1_49 as in boost_1_40. The problem is in the libstd++ that comes with gcc>  4.4
Hi again,
indeed, my ignorance, sorry
> To be honest, I don't fully understand what you are asking here. 
I am doing some FE assembly from some commercial finite element code. If 
you are not familiar with Finite element methods, it is normal that you 
can not understand.

I basically do an operation, with generalized_vector_of_vector, sth like

A( index_array_row, index_array_col ) = A( index_array_row, 
index_array_col ) + b

where b is a matrix of conforming in size with index_arrays. So I add up 
smaller matrices each time.

What I asked: 'are there faster alternatives to this in some other 
library', however in a benchmark paper with MTL, it was shown that the 
above approach of ublas was the fastest for these kinds of additions, I 
do not remember the exact paper location now however.

> Well, I meant that boost::ublas should use its own implementation of "inplace_merge" (copied for example from sgi, if license permits). As I wrote in that message, the main problem would be to sort out potential license issues. But I have no idea what the actual status is. The last thing I remember is that David replied that he wanted to take a look at it.
I see, thanks again
Best wishes,
Umut