67c0ed8518
Fix BenchmarkConnectionMainReadLoop
2018-06-25 14:09:08 +10:00
195695042d
Add requestStrategy 3, which duplicates requests only after a timeout
...
Possible solution for #253 .
2018-06-24 20:04:31 +10:00
e97f487d2e
Rename Config->ClientConfig and change how defaults work
...
This is a very breaking change.
2018-06-16 16:30:04 +10:00
5412eef6da
Fix Count usage in benchmark
2018-06-12 22:47:46 +10:00
cea5584d6b
Track outgoing through a new field on connection, and rework duplicate connection preferencing
2018-06-12 20:14:00 +10:00
ab48d4731c
Include rate limiting and stats in BenchmarkConnectionMainReadLoop
2018-06-11 12:20:51 +10:00
e0c2265cb0
Support sending HaveAll
2018-02-04 19:10:25 +11:00
efecf88d3c
Use flip buffering for connection writing
2018-02-03 15:09:38 +11:00
7a95714df2
Factor out connection.onReadRequest
2018-02-02 19:19:14 +11:00
b0c1f99509
Merge a bunch of stuff into ConnStats and refactor connection.upload
2018-02-02 19:04:56 +11:00
e3c098441a
Add Torrent.setInfo
2018-01-25 17:10:37 +11:00
ed7e3f6ec0
Make Torrent.pendingPieces a priority bitmap in preparation for #220
2018-01-12 12:24:37 +11:00
dd083a4e11
Track completion known to implementation state
...
Addresses #193
2017-10-12 16:09:32 +11:00
b3137b1ede
Make type piece public
2017-09-15 19:35:16 +10:00
13e79039f2
Generate requests, cancels and interest state in the connection writer
2017-08-31 23:48:52 +10:00
986b9093a0
Make connection.writer pull from a buffer
2017-08-31 16:26:45 +10:00
e34681d5d5
Accept racy result in keep alive test
2017-08-16 17:05:34 +10:00
90ca45dd2d
Don't accept pieces with bad indexes
...
Fixes #132
2016-11-22 14:20:48 +11:00
c34234bf34
Split connection.rw into separate Reader and Writer
...
This will make it easier to write hooks for Read and Write separately.
2016-10-10 16:30:51 +11:00
de761fb506
Do chunk pooling at Torrent instead of connection level
2016-10-05 15:57:00 +11:00
39f5d76de4
Add a benchmark for readMainLoop piece decoding
2016-09-11 15:43:57 +10:00
dcfee93f96
Fix race marshalling a bitfield after a Have has also been posted
2016-05-11 23:50:21 +10:00
0fd73396fd
missinggo.Event changed, connection.writeOptimizer changes
2016-05-07 18:56:44 +10:00
bfd0b86441
Tidy up the peer have{,all,none},bitfield handling
2016-03-22 12:07:03 +11:00
08c0c49c4a
Remove internal/pieceordering
2016-01-18 20:11:53 +11:00
06445f2a1e
It's working and the tests are usually passing
...
I still need to handle "prefetch"-style downloading, and some functions haven't been committed to force this issue.
2016-01-18 18:35:14 +11:00
53259397fc
Initialize connection piece priorities lazily, and using a pool
2015-09-26 17:27:35 +10:00
64848a206a
Replace go-check with testify
...
It's muuuuch better.
2015-08-03 16:23:05 +10:00
21fb4933fc
Change the way piece state is exposed to give more detail
2015-06-01 18:22:12 +10:00
3aa78763aa
Use rogpeppe's sortimports to fix this goimports ordering madness
2015-03-26 17:18:08 +11:00
842a32ae0c
Rewrite import paths for migration from Bitbucket
2015-03-20 16:37:44 +11:00
30c67caced
Allow some overlap between piece prioritization classes
...
This probabilistically improves download speeds when doing readaheads
2015-03-20 10:52:55 +11:00
4d510ddca1
Piece priority tests
2015-03-18 18:37:52 +11:00
e4209a2d00
Reorder some imports
2014-12-26 17:17:00 +11:00
7f6987b2a4
Error in connection.writer goroutine wasn't killing the connection
2014-07-18 02:37:33 +10:00
b53e019a6f
Fix broken connection write optimizer test
2014-06-26 17:30:54 +10:00
cfc282ff51
Remove requests from the outbound message queue if cancelled before they're written
...
Only post peer protocol messages to the channel, bytes must be done directly.
This fixes a possible issue where slow responses during handshake could cause
keep alive messages to be sent prematurely.
2014-05-29 01:27:48 +10:00
Matt Joiner