wborgeaud
6dbd39de80
Passing test
2021-05-04 19:56:34 +02:00
wborgeaud
1bf21d17be
Merge branch 'main' into zero-knowledge
2021-05-04 18:02:01 +02:00
wborgeaud
e7c37dfa00
Merge pull request #31 from mir-protocol/fix-subtraction
...
Fixes a bug in the field subtraction
2021-05-04 17:59:23 +02:00
wborgeaud
20dae028d8
fix subtraction
2021-05-04 17:57:59 +02:00
wborgeaud
eb3011b02a
More work on polynomial commitments
2021-05-04 17:48:26 +02:00
wborgeaud
bb8a68e198
Progress on polynomial commitment
2021-05-03 15:17:05 +02:00
wborgeaud
0fa0942981
FRI on coset
2021-04-30 15:07:54 +02:00
Daniel Lubarov
116f92eade
Merge pull request #28 from mir-protocol/mds
...
Add methods for MDS matrices
2021-04-29 07:03:57 -07:00
wborgeaud
c464c038af
Merge pull request #27 from mir-protocol/fri-reduction-arity-contd
...
More work on FRI reduction arity
2021-04-29 08:23:40 +02:00
wborgeaud
fd3e8bcd4c
Minor fixes
2021-04-29 08:18:31 +02:00
Daniel Lubarov
6db30c6db3
Precomputed MDS matrix for CrandallField
2021-04-28 14:42:21 -07:00
Daniel Lubarov
5913fd71b9
Add methods for MDS matrices
...
The defaults are quite slow, but we will override them with fast, precomputed, field-specific matrices; see #4 .
2021-04-28 13:59:58 -07:00
Daniel Lubarov
83354972aa
cargo fmt
2021-04-28 13:57:22 -07:00
wborgeaud
8590407764
Fixes based on PR comments
2021-04-28 22:55:16 +02:00
Daniel Lubarov
aabfbe82cd
Fix GMiMCEvalGate
2021-04-28 13:50:40 -07:00
wborgeaud
f624415a3c
Clippy
2021-04-28 18:43:09 +02:00
wborgeaud
79a8ccd9a0
Working bit-reversed version
2021-04-28 18:38:05 +02:00
Daniel Lubarov
815a294305
Merge pull request #25 from mir-protocol/fix_challenger
...
Fix Challenger's duplicate challenge bug
2021-04-28 09:12:28 -07:00
Daniel Lubarov
ae771bb8bc
Merge pull request #26 from mir-protocol/recursive_gmimc
...
Recursive evaluation of GMiMCGate
2021-04-28 09:05:41 -07:00
Daniel Lubarov
d3ac365667
Recursive evaluation of GMiMCGate
...
If we did it all with `ArithmeticGate`s, the main loop (with ~101 iterations of cubing and a couple adds) would be fairly expensive, so this uses a (much smaller) custom gate called `GMiMCEvalGate` which does all the computations for one iteration of that loop.
2021-04-27 13:16:24 -07:00
Daniel Lubarov
0252d5c762
Fix Challenger's duplicate challenge bug
...
`absorb_buffered_inputs` is called even if the input buffer is empty. In that case it should no-op, but it was instead replenishing the output buffer because of this line:
self.output_buffer = self.sponge_state[0..SPONGE_RATE].to_vec();
Easiest fix is to skip that code if the input buffer is empty.
2021-04-27 09:10:01 -07:00
wborgeaud
d5da6308b5
Merge pull request #24 from mir-protocol/fri-reduction-arity
...
Fri reduction arity
2021-04-27 12:34:20 +02:00
wborgeaud
deb981e97b
More fixes
2021-04-27 09:21:04 +02:00
wborgeaud
187b122c62
Fixes based on Daniel's PR comments.
2021-04-27 08:44:34 +02:00
Daniel Lubarov
cf98d7bccb
Merge pull request #22 from mir-protocol/field_tweaks
...
Tweaks to CrandallField
2021-04-26 11:42:12 -07:00
Daniel Lubarov
460ca64b63
Merge pull request #21 from mir-protocol/perm_arg
...
Port over some code for the permutation argument
2021-04-26 11:41:56 -07:00
wborgeaud
a8c23a357e
Merge branch 'main' into fri-reduction-arity
2021-04-26 19:20:10 +02:00
wborgeaud
f40aba3205
Cleaning and commens
2021-04-26 19:19:27 +02:00
wborgeaud
1786f5e07d
Merge pull request #23 from mir-protocol/montgomery-barycentric
...
Montgomery inversion in computation of barycentric weights
2021-04-26 18:51:28 +02:00
wborgeaud
67aa704f6a
Working reduction arity
2021-04-26 18:24:57 +02:00
wborgeaud
406df34990
Merge from main
2021-04-26 11:00:59 +02:00
wborgeaud
2dfdc39680
More progress on arity
2021-04-26 10:58:27 +02:00
wborgeaud
49c116d7f6
Montgomery inversion in computation of barycentric weights + clean eval_naive method.
2021-04-26 09:13:28 +02:00
Daniel Lubarov
110a7bc6d9
Fill in a few missing field methods
2021-04-25 18:09:43 -07:00
Daniel Lubarov
ffaa9587fe
cargo fmt
2021-04-25 17:16:18 -07:00
Daniel Lubarov
872673702e
Tweaks to CrandallField
2021-04-25 17:14:26 -07:00
Daniel Lubarov
53252af4ba
Forgot to add new file
2021-04-25 17:05:27 -07:00
Daniel Lubarov
aaa0e4aa4a
Port over some code for the permutation argument
...
This is mostly copy/pasted from plonky1, although there are some differences. E.g. in plonky2 virtual targets are not routable, so they're no longer added as partitions.
2021-04-25 17:02:02 -07:00
Daniel Lubarov
5cf8c50abf
Merge pull request #15 from mir-protocol/lagrange_interp
...
Interpolants of arbitrary (point, value) lists
2021-04-24 20:14:09 -07:00
Daniel Lubarov
06bb902f23
Barycentric formula
2021-04-24 20:11:00 -07:00
Daniel Lubarov
035d15bc3d
Interpolants of arbitrary (point, value) lists
...
Closes #10 . This combines Lagrange interpolation with FFTs as mentioned there.
I was previously thinking that all our polynomial encodings might as well just use power-of-two length vectors, so they'll be "FFT-ready", with no need to trim/pad. This sort of breaks that assumption though, as e.g. I think we'll want to compute interpolants with three coefficients in the batch opening argument.
I think we can still skip trimming/padding in most cases, since it the majority of our polynomials will have power-of-two-minus-1 degrees with high probability. But we'll now have one or two uses where that's not the case.
2021-04-24 19:20:18 -07:00
Daniel Lubarov
6c85771ecb
Merge pull request #19 from mir-protocol/prover_use_trees
...
Have the prover use the new MerkleTree API
2021-04-24 11:20:44 -07:00
Daniel Lubarov
a50ba9f590
More unnecessary clones
2021-04-24 11:20:28 -07:00
Daniel Lubarov
6d164adc6a
Have the prover use the new MerkleTree API
...
Before it was storing leaf data and Merkle roots, but nothing in between, since it wasn't yet interacting with intermediate layers (but it will once we hook up the FRI code).
2021-04-24 11:15:11 -07:00
Daniel Lubarov
518470a2d7
Merge pull request #20 from mir-protocol/beta_gamma
...
Properly use the three betas and gammas
2021-04-24 10:54:32 -07:00
Daniel Lubarov
7ffb9cf9b2
Merge pull request #17 from mir-protocol/rescue_bench
...
Make Rescue a bit faster
2021-04-24 10:53:52 -07:00
Daniel Lubarov
d18210d6f1
Merge pull request #16 from mir-protocol/clippy
...
Address some clippy warnings
2021-04-24 10:42:01 -07:00
Daniel Lubarov
03113e85ba
Merge pull request #18 from mir-protocol/no_next_access
...
Remove access to "next" wire & constant values
2021-04-24 10:38:37 -07:00
Daniel Lubarov
4f9aa8879b
Properly use the three betas and gammas
...
... for the three different `Z`s we use. Before I was just using the first value as a temporary thing.
2021-04-23 14:25:24 -07:00
Daniel Lubarov
b18f152c43
Remove access to "next" wire & constant values
...
As discussed, it seems like the batch opening argument will be a significant cost, and we can reduce that cost by not including shifted openings (except for `Z`s which need them).
2021-04-23 13:54:35 -07:00
