[petsc-users] VecView output to HDF5 in 3.12.0 broken ?

[Stefano Zampini]

Actually, after looking at it, that commit is fine. I’m not sure what’s happening here as I’m not an HDF5 expert

> On Oct 12, 2019, at 12:00 AM, Stefano Zampini <stefano.zampini at gmail.com> wrote:
> 
> This is the commit that undo the previous fix https://gitlab.com/petsc/petsc/commit/ccf5ae0b07068d1ada434c93b3759f9651b5ab8c <https://gitlab.com/petsc/petsc/commit/ccf5ae0b07068d1ada434c93b3759f9651b5ab8c>
> 
>> On Oct 11, 2019, at 11:54 PM, Stefano Zampini <stefano.zampini at gmail.com <mailto:stefano.zampini at gmail.com>> wrote:
>> 
>> The size you want to use 268435456, multiplied by 8 (size of double) gives 2147483648, 1 more than the largest 32bit int.
>> Not sure if this is a limitation in how PETSc creates the HDF5 metadata or not, I did not write that code
>> 
>>> On Oct 11, 2019, at 11:35 PM, Sajid Ali via petsc-users <petsc-users at mcs.anl.gov <mailto:petsc-users at mcs.anl.gov>> wrote:
>>> 
>>> 268435456
>> 
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20191012/01f76448/attachment.html>