-
Notifications
You must be signed in to change notification settings - Fork 167
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
Less aggressive inlining. #1857
Merged
Merged
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
This commit is the beginning of reworking our inlining policy to be less aggressive. Currently it inlines all functions called from SOACs that involve arrays in any way. This is not quite enough: we also need to inline anything that contributes to irregularity. Given our downstream compilation restrictions, inlining isn't just about performance - it is about whether we can compile at all. On heston32, the impact of this is pretty promising. Pretty much unchanged performance, but more than 2x faster compilation.
I've realised we will also need to augment the autotuner to be able to tune threshold parameters used in callees. This will require us to expose the call graph of the program in some machine-readable manner. I think the best place to do this is in the manifest. |
This is because constant code generated from functions is low-level bookkeeping things, like the counter arrays used for GPU reductions. These functions might then be used in the normal init code.
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Outdated suggestions cannot be applied.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 commit is the beginning of reworking our inlining policy to be less aggressive. Currently it inlines all functions called from SOACs that involve arrays in any way. This is not quite enough: we also need to inline anything that contributes to irregularity.
Given our downstream compilation restrictions, inlining isn't just about performance - it is about whether we can compile at all.
On heston32, the impact of this is pretty promising. Pretty much unchanged performance, but more than 2x faster compilation.