Skip to content
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

All performance benchmarks should be available via AOT compilation (not just JIT) #5119

Open
steven-johnson opened this issue Jul 15, 2020 · 1 comment

Comments

@steven-johnson
Copy link
Contributor

Currently, all of our performance/ benchmarks are available only via JIT. This is suboptimal:

  • some targets (eg mobile devices) may be incapable or impractical for using JIT (either because of general JIT constraints in the OS or added size/memory of adding libHalide/LLVM into every executable)
  • some targets (eg WASM) have very slow JITs suitable only for correctness

The Makefile currently has a benchmark_apps target that runs some of the apps/ folders as benchmarks (via AOT), but this isn't yet support in CMake.

@shoaibkamil
Copy link
Contributor

This would be a component for #5047.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants