preliminaries for bumping nightly to 2023-08-25

[t-nelson]

Problem

four months worth of nightly lints

Summary of Changes

fix them shits

[t-nelson]

cc/ @AshwinSekar

[ilya-bobyr]

Considering, Ord is implemented for HeaviestSubtreeForkChoice, this can also be

Suggested change

	Some(self.tree_root.cmp(&other.tree_root))
	Some(self.cmp(other))

It is actually what the standard library documentation suggests: https://doc.rust-lang.org/nightly/std/cmp/trait.PartialOrd.html#how-can-i-implement-partialord

[t-nelson]

but comparing the other fields would change semantics, which i'm not trying to do here

[ilya-bobyr]

PartialOrd and Ord implementations must be consistent.
It is part of the trait description:

If Ord is also implemented for Self and Rhs, it must also be consistent with partial_cmp (see the documentation of that trait for the exact requirements).

From https://doc.rust-lang.org/nightly/std/cmp/trait.PartialOrd.html

So, it is actually better to define one in terms of another. Reducing the chance they diverge.
Currently, they are consistent:

solana/core/src/consensus/heaviest_subtree_fork_choice.rs

Lines 201 to 206 in 166f9e2

	#[cfg(test)]
	impl Ord for HeaviestSubtreeForkChoice {
	fn cmp(&self, other: &Self) -> std::cmp::Ordering {
	self.tree_root.cmp(&other.tree_root)
	}
	}

[t-nelson]

yeah that's an ashwin question. don't want to break consensus for the sake of following someone else's rules 😉

[AshwinSekar]

Either approach is fine, they both only compare tree roots. Furthermore this impl is only used for tests, we never compare HeaviestSubtreeForkChoice in consensus.

[ilya-bobyr]

Clippy is really good at finding all these redundancies and unnecessary complications!

[ilya-bobyr]

Considering, Ord is implemented for HeaviestSubtreeForkChoice, this can also be

Suggested change

	Some(self.tree_root.cmp(&other.tree_root))
	Some(self.cmp(other))

It is actually what the standard library documentation suggests: https://doc.rust-lang.org/nightly/std/cmp/trait.PartialOrd.html#how-can-i-implement-partialord

[t-nelson]

Clippy is really good at finding all these redundancies and unnecessary complications!

can't wait to not have to nag about some of these thing in review! 😅

[AshwinSekar]

Looks good, i checked the following:

immutable_deseriaized_packet.rs
unprocessed_transaction_storage.rs
cluster_info_vote_listener.rs
commitment_service.rs
consensus.rs
heaviest_subtree_fork_choice.rs
latest_validator_votes_for_frozen_banks.rs
optimistic_confirmation_verifier.rs
cluster_slot_state_verifier.rs
replay_stage.rs
verified_vote_packets.rs

[AshwinSekar]

Either approach is fine, they both only compare tree roots. Furthermore this impl is only used for tests, we never compare HeaviestSubtreeForkChoice in consensus.