Brean0
commented
3 months ago L-01 - Due to integer division, there are cases where the result may be +/- 1 off. In this example (test_calcRateStableA), you can double reserve[0] and reserve[1] and verify that the test passes.
L-02 - Agreed
L-03 - In Curve's Stableswap Implementation, we can see that _A is set to A * n ** (n - 1) (for a 2 token pool, A * 2), whereas in this implementation, A is directly set. Thus, I don't think this is valid.
L-04 - 1 here can be increased, if a future developer would like to sacrifice tolerance for a lower gas cost on swaps. Thus, <= was used here rather than ==.
L-05 Agreed.
L-06 This is intentional. A well can upgrade its stable well function to use a different A parameter by using a upgradeable well implementation and upgrading the well function.
c4-judge
thebrittfactor
See the markdown file with the details of this report here.