Differentiation between different cargo profiles

[Kailai-Wang]

It's mentioned here that using LTO flag would compile into a more efficient binary at the cost of longer compilation time:
paritytech/polkadot#4311
paritytech/substrate#10608

We might want it to build the releasing binary instead using the default release profile.

[geastwood]

Is it still experimental, or is it for production?

[Kailai-Wang]

Is it still experimental, or is it for production?

For production, at least moonbeam is already using this in their released binaries.
If polkadot/substrate decides to follow, we could do that too.

[ggwpez]

If polkadot/substrate decides to follow, we could do that too.

Already the case, see paritytech/polkadot#4311 (comment)

[Kailai-Wang]

If polkadot/substrate decides to follow, we could do that too.

Already the case, see paritytech/polkadot#4311 (comment)

Thank you for letting us know, will follow up

[Kailai-Wang]

paritytech/polkadot#4778

[github-actions]

❗ This issue is stale because it has been open for 20 days with no activity.
Remove S2-stale label or update it, otherwise this issue will be closed in 5 days.
@litentry/parachain

[Kailai-Wang]

Extended to 2 days as it involves a differentiation between different cargo profiles
see:
https://doc.rust-lang.org/cargo/reference/profiles.html
https://github.com/paritytech/polkadot/blob/master/Cargo.toml

[Kailai-Wang]

locally it took me 78min to compile the litentry-collator binary with production profile