feat: merkle tree proofs

Author: mattrltrent

Cargo.lock

@@ -1,5 +1,5 @@
 [workspace]
-members = ["lib", "lib2", "lib3", "program", "script"]
+members = ["lib", "lib3", "program", "script"]
 resolver = "2"
 
 [workspace.dependencies]

Cargo.toml

@@ -0,0 +1,20 @@
+.PHONY: test-merkle-tree-lib build-program run-program execute-program-gen-core-proof-groth16 execute-program-gen-core-proof-plonk
+
+test-merkle-tree-lib:
+	@cd lib3 && cargo test
+
+build-program:
+	@cd program && cargo prove build
+
+run-program:
+	@cd script && cargo run --release -- --execute --data "a b c d" --index 2
+
+# for these two cmds below, have Docker funning; also, know it might take a few minutes
+#
+# they could also auto-download image: "ghcr.io/succinctlabs/sp1-gnark:v3.0.0"
+#
+# from what I can see... if you don't have enough memory, docker run will fail after about ~10 minutes on my machine
+execute-program-gen-core-proof-groth16:
+	@cd script && RUST_LOG=info cargo run --release --bin evm -- --data "a b c d" --index 2 --groth16
+execute-program-gen-core-proof-plonk:
+	@cd script && RUST_LOG=info cargo run --release --bin evm -- --data "a b c d" --index 2 --plonk

Makefile

@@ -3,10 +3,34 @@
 This is a template for creating an end-to-end [SP1](https://github.com/succinctlabs/sp1) project
 that can generate a proof of any RISC-V program.
 
+# Shorthands
+
+Use the `Makefile` to run the following commands from the root directory of the project:
+
+```sh
+# runs tests on the merkle tree lib
+make test-merkle-tree-lib
+
+# builds the main program (creates and sets `SP1_ELF_merkle-tree-program`)
+make build-program
+
+# runs the main program with dummy data
+make run-program
+
+### these output fixtures for evm-compatible proofs into
+### the ~/contracts/fixtures directory upon succesfully completing
+#
+# generates a core proof (more time-consuming) for groth16 + you need Docker running
+make execute-program-gen-core-proof-groth16
+# generates a core proof (more time-consuming) for plonk + you need Docker running
+make execute-program-gen-core-proof-plonk
+```
+
 ## Requirements
 
 - [Rust](https://rustup.rs/)
 - [SP1](https://docs.succinct.xyz/getting-started/install.html)
+- Docker (for core proof generation)
 
 ## Running the Project
 
@@ -73,7 +97,7 @@ cargo prove vkey --program fibonacci-program
 
 ## Using the Prover Network
 
-We highly recommend using the Succinct prover network for any non-trivial programs or benchmarking purposes. For more information, see the [setup guide](https://docs.succinct.xyz/generating-proofs/prover-network.html).
+We highly recommend using the Succinct prover network for any non-trivial programs or benchmarking purposes. For more information, see ~~the [setup guide](https://docs.succinct.xyz/generating-proofs/prover-network.html).~~ [this setup guide](https://docs.succinct.xyz/docs/generating-proofs/prover-network/key-setup).
 
 To get started, copy the example environment file:
 
@@ -91,6 +115,14 @@ command:
 SP1_PROVER=network SP1_PRIVATE_KEY=... cargo run --release --bin evm
 ```
 
-## Debugging Note
+## Notes
+
+- If you are running this on MacOS M1 aarch64, don't use `zsh` as your shell. Use `bash` instead; `zsh` throws unintelligible errors.
+
+## Next Steps
+
+- Right now `~/lib` contains the old fibonacci program. `/~lib3` is the current refactoring effort to make this work for Merkle Trees.
+- There are some `.expect()` and `.unwrap()` calls in the code that should be handled more gracefully (marked with `// todo ...`s).
+- The `evm.rs` file should be generating fixtures for the EVM-compatible proof (output in `~/contracts/fixtures`), but I think my computer is running out of memory to do so; everything code-wise *seems* correct.
+- To deploy on their [Prover Network](https://docs.succinct.xyz/docs/generating-proofs/prover-network) you need to sign up for beta access! So, it could be worth [signing up for that](https://docs.google.com/forms/d/e/1FAIpQLSd-X9uH7G0bvXH_kjptnQtNil8L4dumrVPpFE4t8Ci1XT1GaQ/viewform) in case it takes a while. They claim this network is the best way to generate proofs for large, complex programs.
 
-If you are running this on MacOS M1 aarch64, don't use `zsh` as your shell. Use `bash` instead; `zsh` throws unintelligible errors.