EIP-2028: Set tx data nonzero byte cost to 16 in Istanbul

[gumb0]

Closes #5690
https://eips.ethereum.org/EIPS/eip-2028

[codecov-io]

Codecov Report

Merging #5691 into master will increase coverage by 0.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master    #5691      +/-   ##
==========================================
+ Coverage   62.98%   62.99%   +0.01%     
==========================================
  Files         350      350              
  Lines       29991    29993       +2     
  Branches     3361     3361              
==========================================
+ Hits        18889    18894       +5     
+ Misses       9882     9881       -1     
+ Partials     1220     1218       -2

[chfast]

Please reference this change in the EIP.

[gumb0]

Posted a link on ethmagicians discussion thread

[holiman]

Regarding implementing this in geth, we see a minor snag with transaction pools -- if they start accepting istanbul-priced txs too early, that's a DoS vector. So we're probably making the pools stateful, and when istanbul hits we flip the switch to use another formula to calculate the intrinsic gas (and thus validity of) a transaction.
I see you did not touch anything in that area -- is that not a concern for aleth ?

[gumb0]

It looks like we don't check intrinsic gas at all when adding to the transaction pool.

@holiman Could you point me to the the area of geth code where this happens?

[holiman]

See https://github.com/ethereum/go-ethereum/blob/b80c840af33aae1ff99ce9879fecc6161a0c1620/core/tx_pool.go#L543, which uses the method here: https://github.com/ethereum/go-ethereum/blob/b80c840af33aae1ff99ce9879fecc6161a0c1620/core/state_transition.go#L79