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

APIv2: cleanup redundant bench tests? #886

Open
dsnet opened this issue Jul 2, 2019 · 2 comments

Comments

Projects
None yet
3 participants
@dsnet
Copy link
Member

commented Jul 2, 2019

CL/183245 added a general purpose benchmark framework for all the various serialization formats.

Do we still need the others?

\cc @neild @cybrcodr

@dsnet dsnet added the api-v2 label Jul 2, 2019

@neild

This comment has been minimized.

Copy link
Contributor

commented Jul 3, 2019

Probably not, although proto/bench_test.go has been interesting for comparing the performance of very specific things (maps, oneofs, etc.) on a very micro level. It might be interesting to add some focused benchmarks for those to the benchmarks directory.

@cybrcodr

This comment has been minimized.

Copy link
Contributor

commented Jul 3, 2019

I just checked out benchmark/bench_test.go. That's neat. Yeah, that can replace encoding/bench_test.go.

Btw, just a note that passing in -v1 to benchmark text marshal/unmarshal may produce the same results depending on which v1 is being used. With the go.mod file checked in, I believe it is using the latest v1 where text marshal/unmarshal are already wrappers to the v2 implementation.

In order to properly get v1 values, one will need to adjust go.mod to point to the real old v1 implementation.

@dsnet dsnet removed the api-v2 label Jul 10, 2019

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.