[petsc-dev] Faulty logic in MatHYPRESetPreallocation?

[Pierre Jolivet]

Hello,
I am not sure the hypre routines are called in the right order in MatHYPRESetPreallocation.
Indeed, attached, you’ll find two Massif logs, one with the current order (massif.out.bad), and the other (massif.out.better) with the attached patch applied.
In the former, you can see at the snapshot #31 that there is a call to hypre_CTAlloc in hypre_AuxParCSRMatrixInitialize (aux_parcsr_matrix.c:163).
This means that hypre is allocating memory as if it was not aware of the number of nonzero per row, even though I supply in my application the two arrays dnnz and onnz.

Am I missing something?

Thanks,
Pierre

-------------- next part --------------
A non-text attachment was scrubbed...
Name: massif.out.bad
Type: application/octet-stream
Size: 334384 bytes
Desc: not available
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20181217/f8290516/attachment-0002.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: massif.out.better
Type: application/octet-stream
Size: 502865 bytes
Desc: not available
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20181217/f8290516/attachment-0003.obj>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: patch-hypre.txt
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20181217/f8290516/attachment-0001.txt>