From: Michael Stevens (mail_at_[hidden])
Date: 2005-05-29 04:31:45
The current CVS-HEAD looks very good with regard to any regressions.
There is just one I don't think I can fix it without access to the compiler!
If any has some time have a look at
for the error.
I recently committted a change to matrix_assign.cpp which I hoped would help
the compiler and should avoid using element proxies in matrix swaps. However
the tru64 still seems to have problems deep inside the functions instantated
for the swap.
-- ___________________________________ Michael Stevens Systems Engineering Navigation Systems, Estimation and Bayesian Filtering http://bayesclasses.sf.net ___________________________________