feat(node): validator gas exemption

[cryptoAtwill]

This PR introduces bottom up signature submission to the chain where txn gas will not be charged for validators.

Closes #1255

---

This change is 

[drahnr]

Reviewable status: 0 of 11 files reviewed, 4 unresolved discussions

---

fendermint/rpc/src/message.rs line 194 at r1 (raw file):

    }

    /// Send a validator message to the fendermint

nit: to fendermint

---

fendermint/vm/interpreter/src/chain.rs line 424 at r1 (raw file):

                }
            },
            ChainMessage::Validator(v) => match v {

JFYI: I don't have enough context yet to judge if this design is good or not

---

fendermint/vm/interpreter/src/validator.rs line 19 at r1 (raw file):

const SOLIDITY_SELECTOR_BYTES: usize = 8;

pub(crate) fn execute_bottom_up_signature<DB: Blockstore + Clone + 'static>(

A doc comment 'd be nice

---

fendermint/rpc/src/tx.rs line 115 at r1 (raw file):

        if calldata.len() < SOLIDITY_SELECTOR_BYTES {
            return Err(anyhow!("invalid validator calldata"));

nit: bail!()
nit: the error message is not actionable, nor does it yield enough info to debug, here we clearly check for length, so something like: received message is {x} bytes, but must be at least {SOLIDITY_SELECTOR_BYTES} for the selector

[cryptoAtwill]

Reviewable status: 0 of 12 files reviewed, 4 unresolved discussions (waiting on @drahnr)

---

fendermint/rpc/src/message.rs line 194 at r1 (raw file):

Previously, drahnr (Bernhard Schuster) wrote…

nit: to fendermint

updated!

---

fendermint/rpc/src/tx.rs line 115 at r1 (raw file):

Previously, drahnr (Bernhard Schuster) wrote…

nit: bail!()
nit: the error message is not actionable, nor does it yield enough info to debug, here we clearly check for length, so something like: received message is {x} bytes, but must be at least {SOLIDITY_SELECTOR_BYTES} for the selector

updated!

---

fendermint/vm/interpreter/src/chain.rs line 424 at r1 (raw file):

Previously, drahnr (Bernhard Schuster) wrote…

JFYI: I don't have enough context yet to judge if this design is good or not

take your time, can discuss over this again next monday as well.

---

fendermint/vm/interpreter/src/validator.rs line 19 at r1 (raw file):

Previously, drahnr (Bernhard Schuster) wrote…

A doc comment 'd be nice

added!

[drahnr]

Reviewable status: 0 of 13 files reviewed, 7 unresolved discussions

---

fendermint/rpc/src/tx.rs line 116 at r3 (raw file):

        if calldata.len() < SOLIDITY_SELECTOR_BYTES {
            return Err(anyhow!(
                "invalid validator calldata, expected at least {} but found only {}",

.. bytes

---

Cargo.toml line 250 at r3 (raw file):

[patch.crates-io]
fvm = { git = "https://github.com/consensus-shipyard/ref-fvm.git", branch = "customize-gas" }

Are we planning to merge that branch into our forked main?

---

fendermint/rpc/src/message.rs line 208 at r3 (raw file):

            .inner
            .transaction(to, method_num, params, value, gas_params);
        let signed = SignedMessage::new_secp256k1(message, &self.sk, &self.chain_id)?;

Can be future refactor, but I really would like to see the following pattern:

let signed_msg: Signed<Message> = Message::new().sign(keystore, pubk, chain_id);
// and reverse
let signed_msg: Signed<Message> = bytes.parse()?;
let message: Message = signed_msg.verify()?;

---

fendermint/testing/materializer/tests/docker_tests/layer2.rs line 22 at r3 (raw file):

const CHECKPOINT_PERIOD: u64 = 2;
const SLEEP_SECS: u64 = 5;
const MAX_RETRIES: u32 = 10;

Can we leave a comment why we need so many retries and such a short checkpoint period?