ferretti at mit.edu
Wed Apr 7 12:16:57 CEST 2010
please note this forum is about quantum-espresso, so please make sure that what
you post here is relevant to the forum.
(otherwise post it directly to me, eg when specifically related to WanT
compilation etc etc)
it is actually difficult to guess what's going on here without any information
about the hardware, configuration (make.sys file) and the input file you are
close to "starting iteration loop" (assuming this means anything), nothing
weird is done, the big part of the memory is already deallocated, and just some
mild use of linear algebra routines is done..
- try to compile the code using the internal blas and lapack libraries
and check if the error is still there.
In particular, I remember I got some run-time problems when linking some
optimized version of the lapack lib (eg acml). The problems where
related to calls to ZGESVD (singular value decompositon). In that case,
linking acml as blas, but the internal version for the lapack lib solved
- are you sure the code is compiled serially ?? (the p4_error below, when
googled, leads to a lot of issues related to MPI)
even if this should be perfectly legal, messing around with libraries
may make things harder to understand
> I used disentangle.x code in WanT package do some calculation, but it stops
> = Starting Iteration loop =
> Initial trial subspace: projected localized orbitals
> p0_13395: p4_error: interrupt SIGx: 6
> what is the problem?
> I run it serial.
> Thanks in advance!
> Southeast University in Nanjing,China
Oxford University, Department of Materials
Parks Road, Oxford OX1 3PH, UK
Tel: +44 (0)1865 612796; Skype: andrea_ferretti
Please, if possible, don't send me MS Word or PowerPoint attachments
Why? See: http://www.gnu.org/philosophy/no-word-attachments.html
More information about the users