[petsc-users] configure crash

[Dharmendar Reddy]

Hello,
         I rerun the configure with metis and parametis, it works now. I
guess something went wrong when i got the configure crash messages and i
saw the error messages related to metis and parmetis. But now it works.  I
will check the ctetgen with the patch you and it would make sense if the
ctetgen works with scalar-type complex, i would expect the data transaction
between petsc and tegen to use PetscReal.

thanks
Reddy


On Fri, Jun 14, 2013 at 1:33 PM, Satish Balay <balay at mcs.anl.gov> wrote:

> On Fri, 14 Jun 2013, Dharmendar Reddy wrote:
>
> > Hello,
> >          I get the Following error when i use ctetgen: (have a look at
> the
> > attached log file, i copied a snippet here).
>
> Perhaps ctetgen would work with compilex. [Matt can confirm]. You can
> do the following to enable it.
>
> diff --git a/config/PETSc/packages/ctetgen.py
> b/config/PETSc/packages/ctetgen.py
> index 71ec5e6..f09d952 100644
> --- a/config/PETSc/packages/ctetgen.py
> +++ b/config/PETSc/packages/ctetgen.py
> @@ -9,6 +9,7 @@ class Configure(PETSc.package.NewPackage):
>      self.functions         = []
>      self.includes          = []
>      self.liblist           = [['libctetgen.a']]
> +    self.complex           = 1
>      self.compileCtetgen    = 0
>      return
>
>
> > Also, metis and parametis give
> > error when i configure with scalar-type=complex.
>
> It should work. What error do you get? Send logs.
>
> Satish
>
> > I was using MUMPS which
> > requires either parmetis or pt-scotch. I was able to cinfigure with
> > ptscotch. At this point, i am just seting up my code, an i do not know
> > which combination for direct solver will give a good result. I was hoping
> > to configure with the required packages and then experiment.
> >
> *******************************************************************************
> >          UNABLE to CONFIGURE with GIVEN OPTIONS    (see configure.log for
> > details):
> >
> -------------------------------------------------------------------------------
> > Cannot use ctetgen with complex numbers it is not coded for this
> capability
> >
> *******************************************************************************
> >   File "./config/configure.py", line 293, in petsc_configure
> >     framework.configure(out = sys.stdout)
> >   File
> >
> "/home1/00924/Reddy135/LocalApps/petsc/config/BuildSystem/config/framework.py",
> > line 933, in configure
> >     child.configure()
> >   File
> >
> "/home1/00924/Reddy135/LocalApps/petsc/config/BuildSystem/config/package.py",
> > line 552, in configure
> >     self.consistencyChecks()
> >   File "/home1/00924/Reddy135/LocalApps/petsc/config/PETSc/package.py",
> > line 38, in consistencyChecks
> >     raise RuntimeError('Cannot use '+self.name+' with complex numbers
> it is
> > not coded for this capability')
> >
> >
> >
> >
> > On Fri, Jun 14, 2013 at 11:29 AM, Satish Balay <balay at mcs.anl.gov>
> wrote:
> >
> > > On Thu, 13 Jun 2013, Dharmendar Reddy wrote:
> > >
> > > > Hello,
> > > >           I rerun the configure, it works. I had to remove the
> ctetgen,
> > > > metis and parametis
> > >
> > > I don't unserstand why you had to remove them.
> > >
> > > > Is there documentation on dependencies for external
> > > > packages ? If not, it will be useful to have a required package list
> for
> > > a
> > > > given external package.
> > >
> > > its currently documented in the appropriate .py file for that package.
> > > [and if the dependency is missing - you get a configure error]
> > >
> > > Satish
> > >
> >
> >
> >
> >
>
>


-- 
-----------------------------------------------------
Dharmendar Reddy Palle
Graduate Student
Microelectronics Research center,
University of Texas at Austin,
10100 Burnet Road, Bldg. 160
MER 2.608F, TX 78758-4445
e-mail: dharmareddy84 at gmail.com
Phone: +1-512-350-9082
United States of America.
Homepage: https://webspace.utexas.edu/~dpr342
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20130614/80e9be2a/attachment.html>