Hi,
I have a restriction of installing petsc on a community cluster only via
package manager (APT). But it needs to be configured with lapack and
parmetis. Is there a way to do this without building from source?
Thank you.
Sincerely,
Prateek Gupta, PhD
Dear all,
I am struggling with the simple following problem : Having a first matrix
B1 of size n1xm1, a second matrix B2 of size n2 x m2, build a matrix M of
size (n1+n2)x(m1+m2) where the blocks B1 and B2 are the "diagonal" of M
(M[0:n1,0:m1]=B1, M[n1:(n1+n2),m1:(m1+m2)]=B2). In my case, the bloc
On Thu, Jul 21, 2022 at 6:28 AM Emile Soutter
wrote:
> Dear all,
>
> I am struggling with the simple following problem : Having a first matrix
> B1 of size n1xm1, a second matrix B2 of size n2 x m2, build a matrix M of
> size (n1+n2)x(m1+m2) where the blocks B1 and B2 are the "diagonal" of M
> (M
On Thu, Jul 21, 2022 at 5:16 AM Prateek Gupta
wrote:
> Hi,
> I have a restriction of installing petsc on a community cluster only via
> package manager (APT). But it needs to be configured with lapack and
> parmetis. Is there a way to do this without building from source?
>
Can you just install
Looks like the Debian package is built with PTScotch, which is a partitioner
similar to ParMETIS that has better license. (ParMETIS has a non-free license
and is flagged as such on Debian.) All PETSc builds have LAPACK. If you get it
from your package manager, you'll be stuck with a somewhat old
Thanks Jed. Yes PTScotch is a better choice. I will try the debian package
On Thu, 21 Jul 2022 at 5:40 PM, Jed Brown wrote:
> Looks like the Debian package is built with PTScotch, which is a
> partitioner similar to ParMETIS that has better license. (ParMETIS has a
> non-free license and is flag
No. Actually there is a severe limitation on number of files in /home due
to less storage, which is why I ran into a wall building from source in
userspace. I could do it as root but I want to avoid building from source
in that case.
On Thu, 21 Jul 2022 at 5:34 PM, Matthew Knepley wrote:
> On Th
Yes. This I could try
On Thu, 21 Jul 2022 at 5:42 PM, Matthew Knepley wrote:
> On Thu, Jul 21, 2022 at 7:11 AM Prateek Gupta
> wrote:
>
>> No. Actually there is a severe limitation on number of files in /home due
>> to less storage, which is why I ran into a wall building from source in
>> user
On Thu, Jul 21, 2022 at 7:11 AM Prateek Gupta
wrote:
> No. Actually there is a severe limitation on number of files in /home due
> to less storage, which is why I ran into a wall building from source in
> userspace. I could do it as root but I want to avoid building from source
> in that case.
>
Thanks a lot for the reply. The context is that the matrix itself will be
used as a matrix projection in the context of multigrid algorithm. So the
final matrix,P, will be used as the interpolation matrix in PCMG routine (
https://www.mcs.anl.gov/petsc/petsc-3.6/docs/manualpages/PC/PCMGSetInterpola
Good afternoon PETSc Development Team, I have been trying to distribute your program as third party library in my CMake Project. Because I do not want to my Linux users to have to install petsc by their own. I just want them to use my final product that uses petsc dependencies. When I execute make
You can try:
ldd your-executable
And copy over all the .so files listed by it [ignoring the system libraries -
that might be present on the remote machine
Try setting LD_LIBRARY_PATH to this location [n the new machine]
and retry 'ldd your-executable' on the new machine - and make sure there
I am compiling, installing and checking 3.17.3, as posted below, following
the usual instructions after each step.
I am getting an error with check.
Is there anything to do about this?
Should I be concerned?
Thanks
$ ./configure --with-cc=mpicc -with-cxx=mpicxx --with-fc=mpif90
--prefix=/home/
Another alternative: you can install PETSc with:
--with-shared-libraries=0
[but some of the dependent libraries might still be shared - if so - you'll
need to figure out static build for them. If they are installed with
--download-pkg option of configure - likely it will be static]
Satish
On
>
hwloc/linux: Ignoring PCI device with non-16bit domain.
Pass --enable-32bits-pci-domain to configure to support such devices
(warning: it would break the library ABI, don't enable unless really
needed).
This message is comping from your MPI.
You can ignore it and use PETSc.
On Thu 21. Jul 2022 at 14:06, Matthew Knepley wrote:
> On Thu, Jul 21, 2022 at 6:28 AM Emile Soutter
> wrote:
>
>> Dear all,
>>
>> I am struggling with the simple following problem : Having a first matrix
>> B1 of size n1xm1, a second matrix B2 of size n2 x m2, build a matrix M of
>> size (n1+n2
export HWLOC_HIDE_ERRORS=2
seems to be the option to prevent these confusing and annoying messages.
> On Jul 21, 2022, at 9:37 AM, san.tempo...@gmail.com wrote:
>
> I am compiling, installing and checking 3.17.3, as posted below, following
> the usual instructions after each step.
> I am gett
17 matches
Mail list logo