[Csdp] Using cblas and clapack with CSDP

[Henri Kjellberg]

Here are the test outputs. They do not seem to iterate at all (whereas
./example does). What is the expected output of example?

g50.out
Graph is of size 50 103
Iter:  0 Ap: 0.00e+00 Pobj:  1.0355339e+04 Ad: 0.00e+00 Dobj:
 0.0000000e+00
Too many line search failures, giving up.
Failure: return code is 9
Primal objective value: 1.0355339e+04
Dual objective value: 0.0000000e+00
Relative primal infeasibility: 5.18e+03
Relative dual infeasibility: 1.02e+01
Real Relative Gap: -1.00e+00
XZ Relative Gap: 7.21e+01
DIMACS error measures: 5.18e+03 0.00e+00 2.60e+02 0.00e+00 -1.00e+00
7.21e+01

Theta1.out
Iter:  0 Ap: 0.00e+00 Pobj:  1.4644661e+04 Ad: 0.00e+00 Dobj:
 0.0000000e+00
Too many line search failures, giving up.
Failure: return code is 9
Primal objective value: 1.4644661e+04
Dual objective value: 0.0000000e+00
Relative primal infeasibility: 7.32e+03
Relative dual infeasibility: 1.02e+01
Real Relative Gap: -1.00e+00
XZ Relative Gap: 7.21e+01
DIMACS error measures: 7.32e+03 0.00e+00 2.60e+02 0.00e+00 -1.00e+00
7.21e+01


On Sun, Nov 7, 2010 at 9:18 PM, Brian Borchers <borchers at nmt.edu> wrote:

> It appears that something has written over workvec5 and has caused a
> memory management error.  It's good that the cygwin libc has enough
> error checking built in to catch this.  I'll have to investigate
> further to see where exactly the problem is occuring.
>
> Have you tried running the csdp solver on a test problem to see
> whether it works without this error?
>
> --
> Brian Borchers                          borchers at nmt.edu
> Department of Mathematics      http://www.nmt.edu/~borchers/
> New Mexico Tech                       Phone: (575) 322-2592
> Socorro, NM 87801                   FAX: (575) 835-5366
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://list.coin-or.org/pipermail/csdp/attachments/20101107/3f3fb48c/attachment.html