Dear Eduardo,
thank you for your suggestion. I would like to do that, but I cannot
find the scalapack provided by siesta. Is it an extra package that I
have to download from the website?
Simon
Eduardo Anglada wrote:
Dear Simon,
There is a problem with digital scalapack, use the one provided by
siesta
Regards,
Eduardo
On 06/03/2007, at 11:48, Simon Dubois wrote:
Dear Siesta Users,
previously, I used a version of siesta.2.0.1 compiled in parallel
with pgi on AMD processors, and it went fine. Now, I'm trying to use
the same version of siesta in parallel on Alpha processors. The
compilation goes well but, when running in parallel, the code
produces a lot of comments that I'm not used to see...
.......
{ 0, 0}: On entry to PZHEGVX parameter number 2 had an
illegal value
{ 0, 1}: On entry to PZHEGVX parameter number 2 had an
illegal value
{ 0, 1}: On entry to PZHEGVX parameter number 2 had an
illegal value
{ 0, 0}: On entry to PZHEGVX parameter number 2 had an
illegal value
{ 0, 1}: On entry to PZHEGVX parameter number 2 had an
illegal value
{ 0, 0}: On entry to PZHEGVX parameter number 2 had an
illegal value
{ 0, 1}: On entry to PZHEGVX parameter number 2 had an
illegal value
.......
Is there sombody who have already experimented these comments?
Any advice will be appreciated....
Simon
