incubating supernova #2

Merged

porcuquine merged 94 commits intolurk-lang:devfrom

hero78119:arecibo_supernova

Aug 18, 2023

Commits on Jul 26, 2023

start SuperNova

wyattbenno777
authored and
hero78119
committed
setup unit test and start program counter

wyattbenno777
authored and
hero78119
committed
experimenting with the RunningClaims selector.

wyattbenno777
authored and
hero78119
committed
start conversion of RecursiveSNARK into NivcSnark

wyattbenno777
authored and
hero78119
committed
theta function needs to be defined. Starting from public param generator.

wyattbenno777
authored and
hero78119
committed
RunningClaims will be a user provide Struct.

wyattbenno777
authored and
hero78119
committed
moving CK out

wyattbenno777
authored and
hero78119
committed
setup public params in a nice way for SuperNova.

wyattbenno777
authored and
hero78119
committed
starting on phi logic

wyattbenno777
authored and
hero78119
committed
adding pci into code

wyattbenno777
authored and
hero78119
committed
move cks out and also start on U_i and pki as described in paper.

wyattbenno777
authored and
hero78119
committed
Start on SuperNova augmented circuit

wyattbenno777
authored and
hero78119
committed
work on augmented circuit2

wyattbenno777
authored and
hero78119
committed
testing pci as output

wyattbenno777
authored and
hero78119
committed
closer to pci fin

wyattbenno777
authored and
hero78119
committed
push notes

wyattbenno777
authored and
hero78119
committed
more details and analysis of implementation method.

wyattbenno777
authored and
hero78119
committed
change to match paper

wyattbenno777
authored and
hero78119
committed
clarified wording

wyattbenno777
authored and
hero78119
committed
leave write up for review.

wyattbenno777
authored and
hero78119
committed
typo

wyattbenno777
authored and
hero78119
committed
clarity

wyattbenno777
authored and
hero78119
committed
inputize progam_counter

wyattbenno777
authored and
hero78119
committed
start on hash of U_i

wyattbenno777
authored and
hero78119
committed
swapping wording out for hash of Fpci' pre-image

wyattbenno777
authored and
hero78119
committed
testing U_i constraints

wyattbenno777
authored and
hero78119
committed
need to rethink how U_i works in the circuit. Might just be a hash.

wyattbenno777
authored and
hero78119
committed
decided on how to represent the running instance U_i in the cicruits.

wyattbenno777
authored and
hero78119
committed
change to U_i is preset amount of circuits.

wyattbenno777
authored and
hero78119
committed
refactor

wyattbenno777
authored and
hero78119
committed
got output of hash supernova from circuit

wyattbenno777
authored and
hero78119
committed
add new X2 for supernova_hash. All tests are a GO!

wyattbenno777
authored and
hero78119
committed
prep for SuperNova circuit sequence checking

wyattbenno777
authored and
hero78119
committed
generics for circuit passing

wyattbenno777
authored and
hero78119
committed
cleanup

wyattbenno777
authored and
hero78119
committed
verify supernova step

wyattbenno777
authored and
hero78119
committed
add supernova hash witness check
hero78119
committed
cleanup and bug fix
hero78119
committed
add Makefile
hero78119
committed
fix bugs and cleanup generics
hero78119
committed
compile pass but test failed due to shape mismatch
hero78119
committed
refactor to fit supernova implementation
hero78119
committed
introduce rom concept
hero78119
committed
constrain sequence execution by memory commitment
hero78119
committed
rom access commitment: program counter manipulation by step circuit
hero78119
committed
code cleanup
hero78119
committed
constrain pc[i]=z_i[x] in step circuit as well
hero78119
committed
remove number of arguments circuit constraint
hero78119
committed
adapt latest main
hero78119
committed
reuse nova nicv::prove for supernova
hero78119
committed
reuse nova runninginstance for supernova
hero78119
committed
more util function to serve supernova
hero78119
committed
disable supernova by default
hero78119
committed
error pruning: arity length check in synthesis function
hero78119
committed
primary circuit folding can be single element
hero78119
committed
optimise supernova proof size from 2*relax_rc1s to relax_rc1s + 1
hero78119
committed
code cosmetics
hero78119
committed
cleanup and reuse compute_digest, fix typo
hero78119
committed
variable rename and cleanup some leftover
hero78119
committed
better naming and cleanup
hero78119
committed
proper naming on r1cs_shape
hero78119
committed
opt memory usage, remove unnessesary clone trait
hero78119
committed
docs decoration
hero78119
committed
isolated trait/lib for supernova
hero78119
committed
expose supernova module public
hero78119
committed
refactor and reuse most of reference
hero78119
committed
optimize pc handling in step circuit, opt running instance clone
hero78119
committed
fix comment format
hero78119
committed
retain constraint in Nova to narrow down the scope
hero78119
committed
clean up comment
hero78119
committed
add supernova bench
hero78119
committed
almost all passed by reference in supernova recursivesnark
hero78119
committed
supernova soundness and clippy fix
hero78119
committed
fix groupname in supernova bench
hero78119
committed
chores: polish reference usage in prove step
hero78119
committed

Commits on Aug 4, 2023

fix bug in recursive snark base case for only success on first running claim
hero78119
committed

Commits on Aug 8, 2023

Commits on Aug 14, 2023

eliminate program_counter from secondary circuit ro, better error handling
hero78119
committed

Commits on Aug 17, 2023