zksolc is a Solidity compiler for ZKsync. All LLVM versions since 2015 fold `(xor (shl 1, x), -1)` to `(rotl ~1, x)` if run with optimizations enabled. Here `~1` is generated as an unsigned 64 bits number (`2^64-1`). This number is zero-extended to 256 bits on EraVM target while it should have been sign-extended. Thus instead of producing `roti 2^256 - 1, x` the compiler produces `rotl 2^64 - 1, x`. Analysis has shown that no contracts were affected by the date of publishing this advisory. This issue has been addressed in version 1.5.3. Users are advised to upgrade and redeploy all contracts. There are no known workarounds for this vulnerability.

zkvyper is a Vyper compiler. Starting in version 1.3.12 and prior to version 1.5.3, since LLL IR has no Turing-incompletness restrictions, it is compiled to a loop with a much more late exit condition. It leads to a loss of funds or other unwanted behavior if the loop body contains it. However, more real-life use cases like iterating over an array are not affected. No contracts were affected by this issue, which was fixed in version 1.5.3. Upgrading and redeploying affected contracts is the only way to avoid the vulnerability.

era-compiler-vyper is the EraVM Vyper compiler for zkSync Era, a layer 2 rollup that uses zero-knowledge proofs to scale Ethereum. Prior to era-compiler-vype version 1.3.10, a bug prevented the initialization of the first immutable variable for Vyper contracts meeting certain criteria. The problem arises when there is a String or Array with more 256-bit words allocated than initialized. It results in the second word’s index unset, that is effectively set to 0, so the first immutable value with the actual 0 index is overwritten in the ImmutableSimulator. Version 1.3.10 fixes this issue by setting all indexes in advance. The problem will go away, but it will get more expensive if the user allocates a lot of uninitialized space, e.g. `String[4096]`. Upgrading and redeploying affected contracts is the only way of working around the issue.