Leo Lara

@ed255 I see that with `diff` mode of the prestate tracer we can only get the changed locations in the storage in their final value after execution of the transaction....

@ed255 sorry, I read this "The prestate mode returns the accounts necessary to execute a given transaction" that seems to suggest that it only return the accounts, but then it...

This is what the call in question returns (diff mode not enabled) for tx [0xbfeec243916e51f5416dce5c4a4fd4623a13e14ce2e42d9b17a9cac58b2c2c2b](https://etherscan.io/tx/0xbfeec243916e51f5416dce5c4a4fd4623a13e14ce2e42d9b17a9cac58b2c2c2b): https://pastebin.com/fiDp9MFi using geth v1.10.26

Basically for each account that it touches it returns a key-value pair list of the storage, with every value that has been touched and its pre-state value for the transacion....

Hence, if I understood the goal correctly, this tracer does not help to produce the same output that the current implementation of `gen_state_access_trace` does.

@ed255 the pastebin doesn't work, here it is a working link: https://gist.github.com/leolara/d3c463bc80ef930522c2d34f3b116618

> At the end we only care about the AccessSet (not the access order). So I think the correct question would be if the prestate tracer could replace the function...

Another thing is that it looks it should not be that difficult to implement a Go native tracer in geth that returns what we want or more. The code for...

@ed255 from looking at AcessSet::from, now I am convinced that we could implement this based on the prestate tracer. If we only need the AccessSet.

@ed255 well, not for all opcodes. In our code we are tracing an access for SELFBALANCE, CODECOPY and CODESIZE, but that opcode is not in https://github.com/ethereum/go-ethereum/blob/0c40df5f28b70b4aafee526bed7575520f3b827b/eth/tracers/native/prestate.go We could do a...