[petsc-dev] obscure changes in TSGetStages_Theta

[Lisandro Dalcin]

On Sun, 24 Jan 2021 at 10:08, Stefano Zampini <stefano.zampini at gmail.com>
wrote:

> You actually do not need a RestoreStages if you use a scratch
>
> Vec stages[2];
>
> in TS_theta, and pass it back with TSGetStages_Theta. I understand PETSc
> philosophy is that every Get should have a Restore method, but is this
> really necessary for TSGetStages?
>


If the only purpose of TSGetStages() is to just give access to other
internal PETSc objects (like KSPGetPC() for example), then there is
absolutely no need for a Restore.
The Get/Restore pattern is not universal, it is only used when you need to
manage internal resources or state.

Unless the semantics of TSGetStages() are going to change to manage extra
internal stage in TS, the proper implementation is the one Stefano
suggested. Add a `Vec stages[2];` in the TSTheta struct, fill it as
appropriate in TSGetStages(), and return the `stages` array to the caller.
Easy and clean, no API changes, no impact in third party code.


-- 
Lisandro Dalcin
============
Senior Research Scientist
Extreme Computing Research Center (ECRC)
King Abdullah University of Science and Technology (KAUST)
http://ecrc.kaust.edu.sa/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20210124/3dfbe786/attachment.html>