-
Notifications
You must be signed in to change notification settings - Fork 9
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
Various optimizations #280
Conversation
Ah! This was with the UDP configuration. With TCP there is a slight speedup (down: 919 Mb/s -> 1.07 Gb/s, up: 599 Mb/s -> 652 Mb/s) |
I accidentally committed and pushed 15ba1ed and eba6e58 to |
When decoding operation return offset and length of the package. We avoid copying the packet only to copy it again without the first byte.
d9b45cc
to
30f2391
Compare
Ah wait the e2e test already fails at the static client configuration. Locally, it initially also failed there for me, but that was due to firewall rules. |
looks fine to me, may revert the e2e changes and redirect again to /dev/null |
We need to write it before computing the hmac!
c1107ea
to
4456378
Compare
|
I cleaned up the git history a bit by merging two commits and dropping the CI debugging commit. It fails on OCaml 5.3~alpha1 due to |
These are mostly optimizations from #223 but also some new ones since we have encrypt_into etc. functions.
main-bench.exe
is built from 6099a4b (unfortunately without --release)As you can see above there are improvements in execution time, but most notably the number of minor words allocations is almost halved. When testing against an openvpn server I didn't observe much of a difference except, curiously, a slowdown in upload speed (from 826 Mb/s to 757 Mb/s). Why that would I don't know.