FEEvaluation: Add some padding between different data fields #15941
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR adds some padding between the
values_dofs,values_quad,gradients_quadandhessians_quadfields ofFEEvaluationData. This can help the caching strategies of processors, in particular set associative caches and directly mapped caches: In case the number of unknowns or data fields is a power of 2 (e.g. polynomial degree 3 with 4 quadrature points per direction), one can get conflict misses inFEEvaluation::get_value()/get_gradient()when the two fields are some fixed amount of bytes apart, e.g. 4096 bytes on 32 kiB L1 caches with 8 ways, and there are sufficiently many. This PR reduces the probability by arranging the fields with some gaps. These gaps are4 * sizeof(VectorizedArray), and thus break powers of 2 except for the tiny sizes.I have another much bigger PR later where I will also rearrange the entries within the gradients, but this change here is independent.