Balazs Komuves ae3c9827d3

speed up NTT by another almost 10% by using optimized short DFT routines

2025-11-05 17:25:19 +01:00

..

cbits

speed up NTT by another almost 10% by using optimized short DFT routines

2025-11-05 17:25:19 +01:00

Class

fixed the (trivial) C NTT bug, and started a refactor towards a unified type-class based interface

2025-11-04 22:17:29 +01:00

Data

fixed the (trivial) C NTT bug, and started a refactor towards a unified type-class based interface

2025-11-04 22:17:29 +01:00

Field

fixed the (trivial) C NTT bug, and started a refactor towards a unified type-class based interface

2025-11-04 22:17:29 +01:00

FRI

create a "bundle" from all the C files (better inlining opportunities for the C compiler); also add a simple benchmark for the C FFT implementation

2025-11-05 17:06:24 +01:00

Hash

incorporate C monolith hash implementation (TODO: fast hashing of flat arrays of field elements)

2025-10-14 20:21:19 +02:00

NTT

create a "bundle" from all the C files (better inlining opportunities for the C compiler); also add a simple benchmark for the C FFT implementation

2025-11-05 17:06:24 +01:00

Outsource

fast C implementation for Goldilocks field + tests for field implementations

2025-10-14 18:09:30 +02:00

FRI.hs

initial import

2025-10-06 00:57:05 +02:00

Hash.hs

initial import

2025-10-06 00:57:05 +02:00

Misc.hs

fixed the (trivial) C NTT bug, and started a refactor towards a unified type-class based interface

2025-11-04 22:17:29 +01:00

NTT.hs

preliminary C implementation of NTT

2025-11-04 10:58:02 +01:00

runi.sh

create a "bundle" from all the C files (better inlining opportunities for the C compiler); also add a simple benchmark for the C FFT implementation

2025-11-05 17:06:24 +01:00

testMain.hs

implement the FRI verifier algorithm (not extensively tested)

2025-10-12 22:40:18 +02:00