SOCIS 16 - Improve iterative methods for sparse linear systems

[Cris]

Dear all,
I made a proposal for the GSOC16 but I was not accepted, then I want to 
propose myself for the SOCIS-16 for the same project "Improve iterative 
methods for sparse linear systems".

My GSOC proposal and the discussion in the octave mailing list is in the 
thread
http://lists.gnu.org/archive/html/octave-maintainers/2016-04/msg00080.html

Since my proposal was rejected for the GSOC, I would like to know if 
there is something wrong, missing or something to do to improve it, if 
it is possible.

If you are interested to see the topics that I studied and worked on, I 
attach my bachelor thesis 
https://drive.google.com/open?id=0B3Wk65-0JfmrN1FFRDIwYU9jMjg
and the codes that I made (inspired from the gmres already implemented 
in Octave) for this thesis
 https://drive.google.com/open?id=0B3Wk65-0JfmrSzJWOGtVU0NpZjA

Actually I'm working to improve the patch that I've already submitted 
(that can be found in the previous thread) and I want to implement the 
check of the input also for the other algorithms mentioned in the 
project: pcr, bicg, bicgstab, cgs and qmr (I think that it will be ready 
in some days).

Best regards,
Cristiano Dorigo