- Apr 02, 2025
-
-
STEVAN Antoine authoredthis should close #17. SACLIN has been moved to [`gitlab.isae-supaero.fr:dragoon/komodo.nu`](https://gitlab.isae-supaero.fr/dragoon/komodo.nu).
-
STEVAN Antoine authored
## usage ```bash nu scripts/check-mirror.nu ...[ https://gitlab.isae-supaero.fr/dragoon/komodo https://github.com/dragoon-rs/komodo main ] ``` will - check the Nushell version and throw a warning if mismatch - add temporary remotes with random names - fetch the temporary remotes - get the revision of the `$branch` for both temporary remotes - clean the temporary remotes - show a message if the two revisions are not the same and how many commits they are different > **Note** > > this does not conflict with #17
-
- Apr 01, 2025
-
-
Added a html header files, an added metadata line in Cargo.toml and some formatting changes in the doc to have Latex formulas in docs.rs. It works locally, should work online
-
- Mar 05, 2025
-
-
STEVAN Antoine authored
this is to avoid having to change the toolchain in the CI everytime. `rust-toolchain.toml` and `.gitlab-ci.yml` will have to be updated in pairs when bumping / changing the Rust toolchain.
-
- Feb 03, 2025
-
-
STEVAN Antoine authored
this is an attempt at making benchmarks easier to work with - `benchmarks run` will read benchmarks from NUON data and run them - `benchmarks plot` will plot benchmark results - the output format will be something - a directory whose name is the hash of the CPU spec and the Komodo commit hash - contains `cpu.json` with the CPU info - contains `komodo.txt` with the Komodo commit hash - contains NDJSON result files >
💡 **Note** > > results will typically be uploaded to https://gitlab.isae-supaero.fr/dragoon/komodo-benchmark-results >💡 **Note** > > this MR goes alongside the [`komodo-benchmark-results@restart`](https://gitlab.isae-supaero.fr/dragoon/komodo-benchmark-results/-/compare/main...restart) branch ## changelog - bump Nushell to 0.101.0 - parallel `$in` => remove useless `let input = $in` when possible, e.g. still required when using the function's `$in` in a `for` loop - `group-by` changed => `group-by x --to-table` will now produce a table with columns `x` and `items` instead of `group` and `items` as in 0.95.0 - add link to results repo - the Nushell benchmarks lib - rename `--force` to `--no-confirm (-y)` - add `--append` - reject columns that GPLT will complain about, e.g. `$.points.k` for the FRI plots - add `--save` to the FRI plot - move the "field" and "curve group" benchmarks from `benchmarks/src/bin/operations/` to `benchmarks/src/bin/` - remove `benchmarks/params/fri.nu` because it's been uniformized with the other methods - rewrite the README - add main function to `benchmarks/` that runs the benchmarks from a NUON record specification - simplify the output of FRI run ## TODO - [x] fix "_atomic operations_" (done in 4f69a1d6) - [x] check that _plotting_ still works ## images ### Field     ### Linear algebra    ### FEC      ### ZK   ### FRI        
-
- Jan 29, 2025
-
-
STEVAN Antoine authored
the idea is to make the installation of Nushell easier to maintain, especially regarding versions, currently pinned to `0.95.0` > successful run on GitHub: [13032726635](https://github.com/dragoon-rs/komodo/actions/runs/13032726635) this is also to allow easier testing locally with the same Nushell version as in the CI, e.g. ```bash # install in `~/.local/bin/` and have multiple versions make install-nu hash=$(/tmp/nu --no-config-file --commands 'version | get commit_hash') nu_bin=$"$HOME/.local/bin/nu-$hash" cp /tmp/nu $nu_bin make NU=$nu_bin show test ``` or ```bash # install in the repo and overwrite each time make NU_DEST=. install-nu make NU=./nu show test ``` # changelog - Makefile - split the global `.PHONY` rule into _atomic_ rules next to each _phony_ rule - define `NU` and `NU_FLAGS` to allow changing which and how Nushell runs - define `NU_ARCH`, `NU_VERSION`, `NU_BUILD` and `NU_DEST` for Nushell installation - tweak the output of `make show` a bit - add `print-%` rules to print `Makefile` variables, e.g. `make print-NU_FLAGS` would print `--no-config-file` - add `make install-nu` to replace the ones from the CIs - GitLab CI - use `make install-nu` - export `PATH` with `make print-NU_DEST` - GitHub CI - use `make install-nu` instead of [github.com:hustcer/setup-nu](https://github.com/hustcer/setup-nu) - export `PATH` with `make print-NU_DEST` and `GITHUB_ENV`
-
- Jan 28, 2025
-
-
STEVAN Antoine authored
mainly fixes a few typos and improves `algebra` and `semi_avid` tests ## changelog - fix some typos and notes in the documentation - test more cases for `algebra::split_data_into_field_elements` and `algebra::merge_elements_into_bytes`, more data lengths and more modulus respectively - remove a useless `::<Vec<_>>` on a `collect` in `fec` - `semi_avid::tests` - refactor `bls12-381` into a constant - write an "_attack_" function to alter a particular block, `attack<F, G>(block: Block<F, G>, c: usize, base: u128, pow: u64) -> Block<F, G>` - pass a list of attacks, i.e. `attacks: Vec<(usize, usize, u128, u64)>`, to `verify_with_errors_template`, use the same as the previous hardcoded one - pass a list of "_recodings_", i.e. `recodings: Vec<Vec<usize>>`, to `verify_recoding_template` and assert the number of blocks, e.g. `vec![vec![2, 3], vec![3, 5]]` means that recoding 2 and 3 together should verify, and same with 3 and 5, and also that there must be at least 6 blocks - pass a list of "_recodings_", i.e. `recodings: Vec<(Vec<Vec<usize>>, bool)>`, to `end_to_end_with_recoding_template` and assert the number of blocks and the number of source shards, e.g. `vec![(vec![vec![0, 1], vec![2], vec![3]], true)]` means that trying to decode with blocks 2, 3 and a recoded block from 0 and 1 should work, and also that there must be at least 4 blocks and at most 3 source shards - pass `k` and `n` to `run_template`
-
- Jan 06, 2025
-
-
STEVAN Antoine authored
this should avoid the following error when trying to rust the LSP in a toolchain which does not have the `rust-analyzer` component and without running `rustup component add rust-analyzer` manually: ``` error: Unknown binary 'rust-analyzer' in official toolchain 'stable-x86_64-unknown-linux-gnu'. ```
-
- Nov 26, 2024
-
-
STEVAN Antoine authored
in order to help catch issues with the various Nushell scripts and modules in the source base, i propose to add a CI script that will check them all. below is an example error, e.g. when introducing a syntax error in `.env.nu`: ``` Error: × Failed to parse content: │ file: .env.nu │ err: Expected keyword. ```
-
STEVAN Antoine authored
## changelog - add more snippets and instructions - add missing imports - update the "atomic operations" section - add a table of contents
-
STEVAN Antoine authored
this replaces the "_nothing to do_" messages from Nushell commands with more explicite errors.
-
- Nov 21, 2024
-
-
This MR adds an option to perform erasure coding using FFT rather than using a matrix. It also adds the field FP128 in the list of curves Note that there is a redundant function `random_loss` into benchmarks/bin/fec.rs and examples/fec.rs
-
STEVAN Antoine authored
FRI protocol from [`dragoon/fri`](https://gitlab.isae-supaero.fr/dragoon/fri) ## changelog - add binary assets to be used as inputs - add `fri` and `fri_test_utils` as local dependencies until [`dragoon/fri`](https://gitlab.isae-supaero.fr/dragoon/fri) becomes public - add `fri` feature and module (see section below for the public definitions) - fix bug in 32bd6566 - bump Rust in e7a2c244 - add a versatile example - add Nushell pipeline to run benchmarks and plot results - add some tests ## `fri` module API ```rust struct Block<F: PrimeField, H: Hasher> ``` ```rust fn evaluate<F: PrimeField>(bytes: &[u8], k: usize, n: usize) -> Vec<Vec<F>> ``` ```rust fn encode<F: PrimeField>( bytes: &[u8], evaluations: Vec<Vec<F>>, k: usize, ) -> Vec<fec::Shard<F>> ``` ```rust fn prove<const N: usize, F: PrimeField, H: Hasher, P>( evaluations: Vec<Vec<F>>, shards: Vec<fec::Shard<F>>, blowup_factor: usize, remainder_plus_one: usize, nb_queries: usize, ) -> Result<Vec<Block<F, H>>, KomodoError> where P: DenseUVPolynomial<F>, for<'a, 'b> &'a P: Div<&'b P, Output = P>, <H as rs_merkle::Hasher>::Hash: AsRef<[u8]>, ``` ```rust fn verify<const N: usize, F: PrimeField, H: Hasher, P>( block: Block<F, H>, domain_size: usize, nb_queries: usize, ) -> Result<(), KomodoError> where P: DenseUVPolynomial<F>, for<'a, 'b> &'a P: Div<&'b P, Output = P>, <H as rs_merkle::Hasher>::Hash: AsRef<[u8]>, ``` ```rust fn decode<F: PrimeField, H: Hasher>(blocks: Vec<Block<F, H>>, n: usize) -> Vec<u8> ``` ## results ### times      ### sizes    
-
STEVAN Antoine authored
see the successful CI on GitHub: https://github.com/dragoon-rs/komodo/actions/runs/11950087413/job/33310881686
-
- Nov 13, 2024
-
-
STEVAN Antoine authored
this MR adds two new badges to the README - the crate (links to the [Komodo crate](https://crates.io/crates/komodo)) - the documentation (links to the [doc](https://docs.rs/komodo/latest/komodo/))
-
STEVAN Antoine authored
this adds three badges to the README: - the latest release (links to the [release page](https://gitlab.isae-supaero.fr/dragoon/komodo/-/releases)) - the GitLab CI status (links to the [GitLab pipeline dashboard](https://gitlab.isae-supaero.fr/dragoon/komodo/-/pipelines)) - the GitHub CI status (links to the [GitHub workflow runs dashboard](https://github.com/dragoon-rs/komodo/actions))
-
DETCHART Jonathan authored
-
- Nov 06, 2024
-
-
-
-
DETCHART Jonathan authored
This MR enables by default `kzg` and `aplonk`. By doing that, both modules should be visible in the doc.
-
- Nov 05, 2024
-
-
-
STEVAN Antoine authored
-
STEVAN Antoine authored
-
STEVAN Antoine authored
-
- Sep 23, 2024
-
-
STEVAN Antoine authored
## changelog - _semi\_avid_, _kzg_ and _aplonk_ examples have been added - the `fs` module has been hidden behind an `fs` feature - the `conversions` module has been properly hidden behind the `test` config feature - the documentation has been completed - some error messages have been improved > **Note** > > the documentation of aPlonK has been left as-is for now
-
- Aug 06, 2024
-
-
STEVAN Antoine authored
`algebra`, `field` and `linalg` were doing extremely similar things before... this MR merges them into a single module `algebra` - old `algebra` and `field` are at the root of the new `algebra` - old `linalg` is now `algebra::linalg` all references to these have been fixed in the rest of the codebase and the features have been tuned to work fine.
-
STEVAN Antoine authored
in !162 and !163, i did only run `cargo ... --all-features` without checking the individual features... this MR adds some `cargo check --features ...` to the `Makefile` and makes sure the imports make sense when compiling a single feature. >
❗ **Important** > > the other notable change here is that `kzg::commit` has been moved to `zk::ark_commit` and is re-exported from `kzg` as `kzg::commit`. -
STEVAN Antoine authored
this adds the aPlonK cryptographic method alongside Semi-AVID and KZG+. ## changelog - new feature `aplonk`: - the `algebra` module is compiled when either `kzg` or `aplonk` features are enabled - `algebra::scalar_product_polynomial` compiles only with `kzg` - the other `algebra::*` functions compile with `aplonk` - `u32_to_u8_vec` has been moved to new `conversions` module which compiles when either `kzg` or `aplonk` features are enabled - new `aplonk` module which compiles only when the `aplonk` feature is enabled - public structures - `Block` - `Commitment` - `SetupParams` - `VerifierKey` - public functions - `setup` - `commit` - `prove` - `verify` - internals - `ipa::Params` - `ipa::Proof` - `ipa::prove` - `ipa::verify` - `polynomial::compute_g` - `transcript::initialize` - `transcript::reset` - `transcript::hash` -
STEVAN Antoine authored
this adds the KZG+ cryptographic method alongside Semi-AVID. ## changelog - the Makefile now uses `--all-features` to compile everything - a new module `algebra` has been added, with tests for all functions - `algebra::scalar_product_polynomial` computes a linear combination of polynomials $(P_i)$ - `algebra::powers_of` computes $n$ successive powers of a number $r$ - a new function `zk::trim` has been added from [`gitlab.isae-supaero.fr:a.stevan/poly-commit@19fc0d4a/src/kzg10/mod.rs#L513-L538`](https://gitlab.isae-supaero.fr/a.stevan/poly-commit/-/blob/19fc0d4ad2bcff7df030c952d09649918dba7ddb/src/kzg10/mod.rs#L513-L538) - a new feature `kzg` has been added, with an optional dependency on [`gitlab.isae-supaero.fr:a.stevan/poly-commit@19fc0d4a`](https://gitlab.isae-supaero.fr/a.stevan/poly-commit/-/tree/19fc0d4ad2bcff7df030c952d09649918dba7ddb) - a new module `kzg` has been added and exposes the following publicly - structures - `Block` - functions - `commit` - `prove` - `verify` - `batch_verify` - tests when block are left as-is and when one of them is corrupted
-
- Aug 01, 2024
-
-
STEVAN Antoine authored
## changelog - `src/main.rs` has been moved to a new crate: `bins/saclin` which stands for **S**emi-**A**VID **CLI** in **N**ushell - dependencies of `komodo` have been fixed - Nushell and Rust tests have been split in the Makefile: by default, only Rust tests will run locally and Nushell tests and examples can be run manually if desired. The CI will still run everything. - the README has been updated - test images have been moved to `assets/` - the majority of the old `./nu-utils/` module have been moved to internals of `./benchmarks/` and imports have been fixed - `cargo.nu` has been moved to `./bins/` and a new `./bins/README.md` mentions it - `./bins/saclin/` has been created and should be a self-contained Rust crate + Nushell module
-
STEVAN Antoine authored
this is a refactor to prepare the addition of other cryptographic methods. ## changelog - moves Semi-AVID code from `lib.rs` to `semi_avid.rs`
-
- Jul 31, 2024
-
-
STEVAN Antoine authored
- 1c9077f6 add documentation to the errors - c388cbbb better FEC doc - 8d58278d doc of `zk.nb_elements_in_setup`
-
- Jul 12, 2024
-
-
STEVAN Antoine authored
this MR is two-fold - it restructures the two main Nushell modules so that they are easier to read and use - it improves the "run" and "plot" modules for the benchmarks ## changelog - `.nushell/` is now renamed to `nu-utils/` - `benchmarks/` is now a valid Nushell module which exports a bunch of modules - `benchmarks linalg`: measure and plot linear algebra operations - `benchmarks setup`: measure and plot trusted setup building - `benchmarks commit`: measure and plot crafting commitments - `benchmarks recoding`: measure and plot the recoding of shards - `benchmarks fec`: measure and plot FEC operations, such as encoding and recoding, and allow combining these results with the pure recoding ones - the submodules of `benchmarks` typically have a `run` and a `plot` command, whith the exception of `benchmarks fec` which has a `run` module and multiple "plot" commands in `benchmarks fec plot` - the "run" commands will create a random temp file by default and ask for confirmation otherwise if the output file already exists, unless `--force` is used - snippetds in `benchmarks/README.md` have been updated
-
STEVAN Antoine authored
## results     
-
- Jul 05, 2024
-
-
STEVAN Antoine authored
see the changelogs of - [Nushell `0.94.0`](https://www.nushell.sh/blog/2024-05-28-nushell_0_94_0.html) - [Nushell `0.94.1`](https://www.nushell.sh/blog/2024-05-30-nushell_0_94_1.html) - [Nushell `0.94.2`](https://www.nushell.sh/blog/2024-06-03-nushell_0_94_2.html) i also removed the `bytes encode` and `bytes decode` commands from `.nushell/binary.nu` because... well they did not work anymore and they were not used anywhere
👀 -
STEVAN Antoine authored
## results >
💡 **Note** > > using the improved plots from !153    -
STEVAN Antoine authored
this adds a new custom field to the `benchmarks::fields` new module and fixes the "linalg" commands to run the benchmarks ## results >
💡 **Note** > > using the data from !156    -
STEVAN Antoine authored
## results  
-
STEVAN Antoine authored
adds a `.env.nu` file to load Nushell modules automatically thanks to the `nuenv` hook from the `nu-hooks` package.
-
STEVAN Antoine authored
- show the log of the degree for high values - don't show the "time" Y label because the units are in the times values - don't rotate the X tick labels ## results  
-
