-
Notifications
You must be signed in to change notification settings - Fork 105
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
Investigate how other types of sections (excluding code and data) in a wasm module contribute to instantiation time #3942
Comments
Local bench run
Test measurement run
|
For this investigation, three additional benches were added to estimate the approximate instantiation weight of sections:
Bench resultsThe table summarize result of several subsequent runs on reference bench machine.
GlobalAs bench results show, the weight of the global section is quite high at Table (with elements)As bench results show, the weight of the table section (with elements) is high at It's probably better to charge for the instantiation of the table section separately because the table could have an arbitrary size, and as a result, it could be abused. Type sectionThe weight of the types section should be comparable to, or close to, the weight of code section instantiation, because no heavy work should be done during the instantiation of these sections, nor memory copying or allocation. As the benchmark results show, the types weight is SummaryFor the instantiation of global and table sections, we should charge separately. Additionally, it may be necessary to limit the number of segments in these sections. For the type section, it's probably better not to charge separately but to charge it as part of the code section (code size). |
Problem to Solve
Investigate how other types of sections (excluding code and data) in a wasm module contribute to instantiation time.
Section like:
Possible Solution
Investigate by writing instantiation module benches for these sections.
The text was updated successfully, but these errors were encountered: