issues
search
TritonVM
/
triton-vm
Triton is a virtual machine that comes with Algebraic Execution Tables (AET) and Arithmetic Intermediate Representations (AIR) for use in combination with a STARK proof system.
https://triton-vm.org
Apache License 2.0
247
stars
37
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Implement Range for OpStack
#338
cyberbono3
opened
2 weeks ago
1
Optionally produce deterministic proofs
#337
jan-ferdinand
closed
1 week ago
0
test: Verify randomizers' large Hamming distance
#336
aszepieniec
closed
1 month ago
1
Use randomness more efficiently
#335
jan-ferdinand
closed
2 weeks ago
5
Optionally derive STARK prover's randomness deterministically
#334
jan-ferdinand
closed
1 week ago
0
Introduce versioning for proofs
#333
jan-ferdinand
closed
1 month ago
1
chore: Bump dependency indexmap
#332
Sword-Smith
closed
1 month ago
0
Memory-Efficient Code Path is Not Automatically Triggered
#331
aszepieniec
opened
1 month ago
0
Drop Cached Polynomials
#330
aszepieniec
closed
1 month ago
7
Terminology: Disambiguate "Table"
#329
aszepieniec
opened
1 month ago
0
`Display` on `VMError` should also show jumpstack trace
#328
Sword-Smith
closed
2 weeks ago
0
Thv/bfe xfe arithmetic in tasm code
#327
Sword-Smith
closed
1 month ago
0
Thv/faster compilation
#326
Sword-Smith
closed
1 month ago
1
Thv/degree lowering hardening
#325
Sword-Smith
closed
1 month ago
11
Improve ergonomics of failing instruction `assert` and `assert_vector`
#324
jan-ferdinand
opened
2 months ago
4
Break cyclic build dependency
#323
jan-ferdinand
closed
2 months ago
0
Implement `Index<Range<…>>` for `OpStack`
#322
jan-ferdinand
opened
2 months ago
0
Introduce versioning for proofs
#321
jan-ferdinand
closed
1 month ago
0
`MemoryRegion::is_disjoint_from` can overflow on `u64` addition
#320
Sword-Smith
closed
2 months ago
0
Performance of functions vs inlining
#319
chancehudson
closed
3 months ago
9
Immediate memory opcodes
#318
chancehudson
opened
3 months ago
4
Change variables in fill_program_hash_trace function
#317
Einstellung
closed
3 months ago
1
add comments for fill_program_hash_trace function
#316
Einstellung
closed
3 months ago
3
Question about memory performance/safety
#315
chancehudson
closed
3 months ago
7
feat: Dynamic Address for Out-of-Domain Row in TASM AIR Evaluation
#314
aszepieniec
closed
3 months ago
0
feat: New Instruction `addi` "Add-Immediate"
#313
aszepieniec
closed
3 months ago
0
u32-table row count grew a lot with 0.42-alpha 6
#312
Sword-Smith
opened
3 months ago
1
request: Drop wrapper type `PublicInput`
#311
aszepieniec
closed
2 months ago
3
request: Drop generic type argument from `Program::hash`
#310
aszepieniec
closed
3 months ago
0
How to use initial ram state, `divine_sibling` merkle tree path
#309
chancehudson
closed
4 months ago
0
Memory opcode stack layout is incorrect
#308
chancehudson
closed
4 months ago
8
Output values are multiplied by `2**32-1`
#307
chancehudson
closed
4 months ago
2
subtraction/negation opcode
#306
chancehudson
closed
4 months ago
2
feat: triton-cli prove/verify, proof binary serialization
#305
chancehudson
opened
4 months ago
12
Introduce instruction `merkle_step_mem`
#304
jan-ferdinand
closed
2 months ago
0
Change `sponge_absorb_mem` to be used in combination with `recurse_or_return`?
#303
Sword-Smith
opened
5 months ago
0
Change `xy_dot_step` instructions for use in combination with `recurse_or_return`
#302
Sword-Smith
opened
5 months ago
0
Multi-Claim Proofs
#301
aszepieniec
opened
5 months ago
1
Bug in documentation of dot_step instructions
#300
Sword-Smith
closed
5 months ago
0
Add prove verify command
#299
cyberbono3
closed
3 months ago
1
Drop Duplicate Consts
#298
aszepieniec
closed
2 months ago
1
Introduce instructions `pick` + `i` and `place` + `i`
#297
jan-ferdinand
closed
2 months ago
8
Start VM with initialized Sponge state
#296
jan-ferdinand
opened
5 months ago
0
Halve combination codeword checks
#295
Sword-Smith
closed
5 months ago
0
feat: Parallelize table extension
#294
aszepieniec
closed
5 months ago
2
perf: Parallelize calculation of deep codeword inner product
#293
Sword-Smith
closed
5 months ago
1
perf: Parallelize evaluation-part of quotient-LDE
#292
Sword-Smith
closed
5 months ago
0
deps: Bump twenty-first revision
#291
aszepieniec
closed
5 months ago
1
Fix #289 by changing prover's evaluation domain for deep update
#290
Sword-Smith
closed
5 months ago
0
Proving fails for FRI expansion factors (strictly) larger than 4
#289
Sword-Smith
closed
5 months ago
0
Next