hwloc is giving this warning on your machine:
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).
The PETSc results are fine, so you can use your
https://petsc.org/release/faq/#what-does-the-message-hwloc-linux-ignoring-pci-device-with-non-16bit-domain-mean
> On Oct 10, 2022, at 11:54 PM, Sijie Zhang wrote:
>
> Hi,
>
> When I try to install petsc on my PC and run the make all check commend it
> has the following error. Can you help m
On Tue, Oct 11, 2022 at 8:52 AM Barry Smith wrote:
>
>
> https://petsc.org/release/faq/#what-does-the-message-hwloc-linux-ignoring-pci-device-with-non-16bit-domain-mean
Ah. Obscure environment variables, not present in the error message, are
always the best way to deal with this. Stay classy, h
Alternative: build/use mpich without hwloc. i.e:
--download-mpich=1 --with-hwloc=0
Satish
On Tue, 11 Oct 2022, Matthew Knepley wrote:
> On Tue, Oct 11, 2022 at 8:52 AM Barry Smith wrote:
>
> >
> >
> > https://petsc.org/release/faq/#what-does-the-message-hwloc-linux-ignoring-pci-device-with-no
On Mon, Oct 10, 2022 at 10:41 PM Mike Michell wrote:
> Hi, I was wondering if there is any comment on the example file that I can
> refer to.
>
I see the problem. Local2Local is not implemented for Plex. I thought we
had this automated, but it was only
coded for DMDA. It is a fairly mechanical t
Thank you for the reply and checking.
Indeed, it seems that local-to-local halo is still not implemented for
DMPlex. But I believe this is a very required feature for large 3D
simulations with DMPlex.
Would you mind if I ask for any estimated timeline to fix this issue and
put it on the official ve
Hello,
I have a code that handles a PETSc Vec on many procs and I would like to use
VecScatterCreateToZero to have all elements of this vector on a single proc,
call VecGetArrayRead on this proc to get the corresponding array to carry out
some diagnostics.
Unfortunately, what I noticed is tha
You need to first convert the Vec to the "natural" ordering and then bring
that vector down to one rank. Something like
DMDACreateNaturalVector(dm,&n);
DMDAGlobalToNaturalBegin/End(dm, n);
> VecScatterCreateToZero(n, &scat, &Xseq);
> VecScatterBegin(scat, n Xseq, INSERT_VALUES, SCATTER_FO