-
Notifications
You must be signed in to change notification settings - Fork 1k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add optimized linear system assembly for finite volume computations #25745
Conversation
1cd22b0
to
0405e4b
Compare
743e1a9
to
a498165
Compare
e9cbe08
to
117deda
Compare
dce8a94
to
fb0e041
Compare
fb0e041
to
cc6e2fe
Compare
Job Documentation on adbef30 wanted to post the following: View the site here This comment will be updated on new commits. |
f815da4
to
6c962f1
Compare
Job Coverage on adbef30 wanted to post the following: Framework coverage
Modules coverageHeat transfer
Navier stokes
Solid mechanics
Thermal hydraulics
Full coverage reportsReports
Warnings
This comment will be updated on new commits. |
ccc1528
to
545df81
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Just through the first two files. Very nice design page. I will continue reviewing tomorrow
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
beginning to glance through the code, I haven't run into big concerns yet. A couple thoughts I do have:
- There seems like a fair amount of dead, unused code in
LinearSystem
. Can we remove some of it or test some of it or both? - I noticed that the majority of uncovered code is from overrides of methods in
MooseVariableField
that simply aren't relevant to your new variable. Would it make sense to create another class, perhapsMooseVariableNonlinearField
or something like that, that sits in betweenMooseVariableField
andMooseVariableFE/FV
that has all those pure virtuals that are irrelevant to your class?
Another idea would be to inherit directly from |
Well, I would like to keep using variable-based interfaces (throught the FE problem and variablke interface) but I will definitely think about how to make this part cleaner! Yeah |
You may need to make new interfaces. But I think a design in which we don't have all these erroring overrides would lead to a lot of good cleanup. It would likely make it impossible to use a linear variable in a place where only a nonlinear variable will work and visa versa |
This member can be used in downstream objects which use functor interfaces and hence don't care why type of field variable the variable is
00381b6
to
adbef30
Compare
Refs #25722
Major issues to fix:
MooseVariable
inheritance/interfaces for linear variables.