IRC meeting summary for 2016-02-18
- Reorg performance of SequenceLocks checks
Note: The meeting was a short one as some developers were participating in the Bitcoin Roundtable.
Btcdrak suggested arranging open source licenses of the Jetbrains IDEs for C++ and Python. The maintainer (wumpus) needs to apply.
feefilter P2P message
The concept of a limited mempool was introduced in Bitcoin Core 0.12 to provide
protection against attacks or spam transactions of low fees that are not being
mined. Currently there is a reject message which
allows informing peers about insufficient fees, but only on per transaction
feefilter message allows a node to inform its peers about the
minimum transaction fee rate it is willing to accept so that its peers can skip
relaying nonconforming transactions.
Wumpus hasn’t looked at the the fee filter so it will be discussed on the next meeting.
Review Implement “feefilter” P2P message #7542
Reorg performance of SequenceLocks checks
BIP 68 Consensus-enforced transaction replacement signaled via sequence numbers.
BIP 68 changes the meaning of the previously unused sequence number field to a relative locktime.
SequenceLocks functions are used to evaluate sequence lock times or heights per BIP 68.
Checking sequence locks to determine whether the transaction is valid requires looking up the heights of all its inputs. In a reorg, as it stands now, this will require reevaluating the inputs of every single transaction in the mempool. PR #7187 attempts to cache for each transaction the block hash of the latest block containing an input which had a sequence lock. In the event of a reorg, if that hash is still on the chain, you know the previously calculated height and time (also cached) are still valid. This means that ideally most inputs won’t need to be reevaluated.
There was discussion whether to backport this to 0.12 and 0.11. Due to all the changes to the mempool in 0.12 it will probably be non-trivial to backport that optimization to 0.11, which already is way slower.
Review/test Keep reorgs fast for SequenceLocks checks #7187
wumpus Wladimir J. van der Laan
morcos Alex Morcos
paveljanik Pavel Janik
sdaftuar Suhas Daftuar
shea256 Ryan Shea
19:08:51 <morcos> is the topic 7187 now?
19:09:22 <wumpus> we dont' have a topic yet :)
19:09:32 * btcdrak makes wumpus a coffee