highperfocused/bitcoin
1
0
Fork 0
mirror of https://github.com/bitcoin/bitcoin.git synced 2026-01-19 06:43:45 +01:00
Code Issues Packages Projects Releases Wiki Activity

net_processing: indicate more work to do when orphans are ready to reconsider

Browse Source 
When PR#15644 made orphan processing interruptible, it also introduced a
potential 100ms delay between processing of the first and second newly
reconsiderable orphan, because it didn't check if the orphan work set
was non-empty after invoking ProcessMessage(). This adds that check, so
that ProcessMessages() will return true if there are orphans to process,
usually avoiding the 100ms delay in CConnman::ThreadMessageHandler().
This commit is contained in:
Anthony Towns
2022-12-23 00:21:10 +10:00
parent ecb0a3e425
commit c58c249a5b
3 changed files with 21 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
src/net_processing.cpp
View File

@@ -4902,6 +4902,12 @@ bool PeerManagerImpl::ProcessMessages(CNode* pfrom, std::atomic<bool>& interrupt
LOCK(peer->m_getdata_requests_mutex);
if (!peer->m_getdata_requests.empty()) fMoreWork = true;
}
// Does this peer has an orphan ready to reconsider?
// (Note: we may have provided a parent for an orphan provided
// by another peer that was already processed; in that case,
// the extra work may not be noticed, possibly resulting in an
// unnecessary 100ms delay)
if (m_orphanage.HaveTxToReconsider(peer->m_id)) fMoreWork = true;
} catch (const std::exception& e) {
LogPrint(BCLog::NET, "%s(%s, %u bytes): Exception '%s' (%s) caught\n", __func__, SanitizeString(msg.m_type), msg.m_message_size, e.what(), typeid(e).name());
} catch (...) {