Blas Rodriguez Irizar
Blas Rodriguez Irizar
@mina86 Went through your comments. I'm not sure how to re-use the iterator logic with the current code. If you still want to go that way, I'd ask you to...
@mina86 **defeated** We tried, and tried... but honestly cannot figure this out. Our latest state of mind (a dump) ```diff diff --git a/core/store/src/trie/iterator.rs b/core/store/src/trie/iterator.rs index 030608390..4fa39876a 100644 --- a/core/store/src/trie/iterator.rs +++...
Moved to #7509
(from https://github.com/near/NEPs/pull/364#issuecomment-1238147491) > Second, even if the length is correct, the key or signature might still be considered invalid due to this conditions: https://docs.rs/ed25519-dalek/latest/src/ed25519_dalek/public.rs.html#142 https://docs.rs/ed25519/latest/src/ed25519/lib.rs.html#302 In the case of invalid...
> tests failures seem legit When applying the following diff ```diff diff --git a/runtime/near-vm-logic/src/tests/miscs.rs b/runtime/near-vm-logic/src/tests/miscs.rs index aca82c686..06ce87204 100644 --- a/runtime/near-vm-logic/src/tests/miscs.rs +++ b/runtime/near-vm-logic/src/tests/miscs.rs @@ -998,7 +998,7 @@ fn test_ed25519_verify() { assert_eq!(result,...
> Thank you for the update @blasrodri! Can you include in the PR's description the performance impact of going from `blake2` crate to `blake2-rfc`? Oh it's just because i checked...
`blake2` seems to be faster than `blake2-rfc` ```rust use blake2::{ digest::{Update, VariableOutput}, Blake2bVar, }; use criterion::{black_box, criterion_group, criterion_main, Criterion}; fn blake2_blake2b_256(data: &[u8]) -> Vec { // Blake2bVarCore::new_with_params(salt, persona, key_size, output_size)...
> > `blake2` seems to be faster than `blake2-rfc` > > ```rust > > use blake2::{ > > digest::{Update, VariableOutput}, > > Blake2bVar, > > }; > > use criterion::{black_box,...
> Sweet! I forgot to ask the most obvious question: are these numbers for `x86_64`? Yes ^_^ ``` blas@f1:~$ uname -m x86_64 ```
> yeah, i think you're right that the waiter should be in an `UnsafeCell`. this should actually be pretty straightforward to fix, i think. good catch, @kprotty! Tried following your...