Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
33 commits
Select commit Hold shift + click to select a range
13badc1
feat: compress aggregate Ix certificates through FRI and recursive KZG
arthurpaulino Oct 9, 2026
647fb8f
feat(compression)!: replace local setup with bundled verifying keys
samuelburnham Oct 27, 2026
72b2449
feat(compression): CUDA KZG proving on multi-stark's shared backend
samuelburnham Oct 27, 2026
7fe38c1
feat(compression): BLS12-381 PLONK wrapper
samuelburnham Oct 27, 2026
3364f9a
feat(compression): BN254 curve, Ignition setup and CUDA PLONK
samuelburnham Oct 27, 2026
1419b96
perf(proof-compression): hoist query-invariant alpha work out of the …
samuelburnham Oct 27, 2026
bd2be70
docs(compression): document stage-collapse sizing
samuelburnham Oct 27, 2026
d100bd2
refactor(compression): name the verifying key as SP1 does
samuelburnham Oct 27, 2026
26fbb00
perf(proof-compression): widen emulated-field range limbs to 20 bits
samuelburnham Oct 27, 2026
a2e53a0
feat(compression): make BN254 the default KZG curve
samuelburnham Oct 27, 2026
1bee863
fix(compression): refresh optimized BN254 PLONK verification
samuelburnham Oct 27, 2026
2bf1b93
perf(plonkish): pin multi-stark with lazy byte nibbles
samuelburnham Oct 27, 2026
b9f6b82
fix(proof-compression): repair the ported FRI test and isolate the tr…
samuelburnham Oct 27, 2026
5f52c06
style(proof-compression): satisfy clippy in the CUDA tests
samuelburnham Oct 27, 2026
f015e89
refactor(compression): remove BLS12-381 support
samuelburnham Oct 28, 2026
73f0653
chore(deps): pin multi-stark to merged main
samuelburnham Oct 28, 2026
6daf24c
Merge remote-tracking branch 'origin/main' into kzg-bn254
samuelburnham Oct 28, 2026
cfc3520
Fix CI
samuelburnham Oct 28, 2026
ca8ed5b
refactor(plonk): use stock gnark and remove unused tooling
samuelburnham Oct 28, 2026
22160e9
feat(compression): restore fixed-root normalization of aggregate roots
samuelburnham Oct 29, 2026
a5ddac4
feat(compression): add BN254 Groth16 terminal backend
samuelburnham Oct 29, 2026
7b9a7f2
refactor(compression): embed gnark in the Rust terminal worker
samuelburnham Oct 29, 2026
caa4d9b
feat(compression): run compression inside ix and drop the worker
samuelburnham Oct 29, 2026
2ccaf35
docs(compression): describe compression inside ix
samuelburnham Oct 29, 2026
6ee80ff
feat(compression): bundle verifying keys 0.1 and derive them without …
samuelburnham Oct 29, 2026
9e46f7e
Merge remote-tracking branch 'origin/main' into kzg-bn254
samuelburnham Oct 29, 2026
c6c6cb9
refactor(compression): reuse stages by their proofs and clean up
samuelburnham Oct 29, 2026
d893a81
refactor(compression): use the gnark Ignition SRS and the Ix mirror
samuelburnham Oct 30, 2026
8c6ac3b
feat(compression): add ix compress setup over an in-place fixed cache
samuelburnham Oct 30, 2026
84d2997
refactor(compression): consolidate modules, tests and tooling
samuelburnham Oct 30, 2026
5d9551a
build: let Cargo own the Rust build and drop GPU/PLONK CI
samuelburnham Oct 30, 2026
c7e0f4d
chore(deps): pin multi-stark to main 3515632
samuelburnham Nov 2, 2026
380d76d
ci: limit cuda-compile to CUDA compile and link
samuelburnham Nov 2, 2026
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 8 additions & 1 deletion .cargo/config.toml
Original file line number Diff line number Diff line change
@@ -1,2 +1,9 @@
[build]
rustflags = ["-Ctarget-cpu=native"]
rustflags = ["-Ctarget-cpu=native"]

# Lean's bundled toolchain links against glibc stubs older than 2.38, whose
# headers in C23 mode, the default of recent GCC, redirect `strtol` and its
# relatives to `__isoc23_*` symbols the stubs lack. C17 keeps symbols the link
# can resolve.
[env]
CFLAGS = "-std=gnu17"
12 changes: 6 additions & 6 deletions .github/workflows/ci.yml
Original file line number Diff line number Diff line change
Expand Up @@ -142,9 +142,8 @@ jobs:
image: nvidia/cuda:13.3.1-devel-ubuntu26.04
env:
# This job validates feature plumbing and the Rust-staticlib-to-Lean
# link without paying for a GPU runner. Multi-stark CI compiles the
# complete supported cubin set; one representative architecture is
# sufficient for this downstream integration gate.
# link without paying for a GPU runner. One explicit architecture
# is sufficient for this downstream integration gate.
MULTI_STARK_CUDA_ARCHS: "80"
steps:
# No sccache here yet. Once the GPU benchmark lands after #644, wire
Expand Down Expand Up @@ -177,9 +176,10 @@ jobs:
auto-config: false
use-github-cache: false
- name: Compile CUDA Rust targets
run: cargo check --release --locked --workspace --all-targets --features ix-ffi/parallel,ix-ffi/cuda
- name: Link a CUDA-enabled Lean executable
run: IX_CUDA=1 lake build bench-typecheck
run: |
cargo check --release --locked --workspace --all-targets --features ix-ffi/parallel,ix-ffi/cuda,ix-proof-compression/cuda
- name: Link CUDA-enabled Lean executables
run: IX_CUDA=1 lake build ix bench-typecheck

lints:
name: Lint workflows
Expand Down
2 changes: 1 addition & 1 deletion Benchmarks/Typecheck.lean
Original file line number Diff line number Diff line change
Expand Up @@ -430,7 +430,7 @@ def runTypecheckCmd (p : Cli.Parsed) : IO UInt32 := do
IO.eprintln "error: --join requires singleton CheckEnv shards; drop --skip-deps"
return Ix.Benchmark.Results.exitUsage
-- Off by default; CI passes --texray explicitly.
let useTexray := p.hasFlag "texray"
let useTexray ← TracingTexray.enabled (p.hasFlag "texray")
let useInterp := p.hasFlag "interp"
if join && useInterp then
IO.eprintln "error: --join requires the native singleton-shard prover; drop --interp"
Expand Down
Loading
Loading