- 03 Jun, 2020 5 commits
-
-
Dirk McCormick authored
-
Dirk McCormick authored
-
Dirk McCormick authored
-
Dirk McCormick authored
-
Dirk McCormick authored
-
- 02 Jun, 2020 1 commit
-
-
dirkmc authored
* feat: total wants gauge * fix: in gauges count wants regardless of which peers they're sent to * fix: want block gauge calculation * refactor: simplify peermanagerwants
-
- 21 May, 2020 1 commit
-
-
Steven Allen authored
Instead of tracking offsets, just create a "new" slice starting with the broadcast cancel slice. Under the covers, this will just use the same memory over and over.
-
- 20 May, 2020 1 commit
-
-
Dirk McCormick authored
-
- 19 May, 2020 1 commit
-
-
Dirk McCormick authored
-
- 07 May, 2020 1 commit
-
-
Dirk McCormick authored
-
- 01 May, 2020 2 commits
-
-
dirkmc authored
* fix: use one less go-routine per session * fix: send cancel when GetBlocks() is cancelled (#383) * fix: send cancel when GetBlocks() is cancelled * fix: in SessionManager shutdown nil out sessions * fix: sessionWantSender perf * make sessionWantSender.SignalAvailability() non-blocking * Refactor SessionInterestManager (#384) * refactor: customize SessionInterestManager * refactor: SessionInterestManager perf
-
Dirk McCormick authored
-
- 30 Apr, 2020 3 commits
-
-
Dirk McCormick authored
-
Dirk McCormick authored
-
Dirk McCormick authored
-
- 29 Apr, 2020 1 commit
-
-
Dirk McCormick authored
-
- 23 Apr, 2020 4 commits
-
-
Dirk McCormick authored
-
Dirk McCormick authored
-
Dirk McCormick authored
-
Dirk McCormick authored
-
- 22 Apr, 2020 2 commits
-
-
Dirk McCormick authored
-
Steven Allen authored
* feat: small optimizations * feat: move broadcast wantlist into the peermanager This deduplicates some state and allows us to do less book-keeping for broadcast wants. We should probably rename the PeerManager to the WantManager and rename the WantManager to something else. * fix: lint warnings
-
- 21 Apr, 2020 3 commits
-
-
dirkmc authored
* refactor: add reverse index to peerWantManager to speed up cancels * refactor: in peerWantManager use ForEach instead of allocating lists
-
Steven Allen authored
-
Steven Allen authored
* fix: minimize time holding wantlist lock Instead of holding the lock the entire time we prepare a message, hold the lock while we retrieve the wantlist entries, process the entries without the lock, retake the lock, then mark entries as sent. This means: 1. We never sort entries while holding the lock. 2. We allocate exactly three times while holding the lock (once per entry list). * fix: address code review
-
- 17 Apr, 2020 3 commits
-
-
Dirk McCormick authored
-
Dirk McCormick authored
-
Dirk McCormick authored
-
- 16 Apr, 2020 1 commit
-
-
Dirk McCormick authored
-
- 15 Apr, 2020 2 commits
-
-
Dirk McCormick authored
-
Dirk McCormick authored
-
- 14 Apr, 2020 1 commit
-
-
Steven Allen authored
Instead of copying these slices, we can just reuse them.
-
- 13 Apr, 2020 4 commits
-
-
Dirk McCormick authored
-
Dirk McCormick authored
-
Dirk McCormick authored
-
Dirk McCormick authored
-
- 11 Apr, 2020 1 commit
-
-
Steven Allen authored
In case we're sending a _lot_ of wants: * Prioritize cancels. * Then targeted wants. * Finally broadcast wants.
-
- 10 Apr, 2020 2 commits
-
-
Dirk McCormick authored
-
Steven Allen authored
-
- 03 Apr, 2020 1 commit
-
-
Dirk McCormick authored
-