253 commits

Author	SHA1	Message	Date
Charles Kerr	49f46dcddb	(trunk libT) revert r7548, which broke very low speed download limits.. the simplified peer-msgs parsing didn't distinguish between piece & raw data until the piece was done downloading.	2009-01-02 23:28:57 +00:00
Charles Kerr	86d3e70121	(trunk libT) avoid some unnecessary memory fragmentation... for composited objects that have a tr_ratecontrol, contain the it directly rather than a pointer to one allocated elsewhere on the heap.	2009-01-02 20:42:35 +00:00
Charles Kerr	199e38dd66	(trunk libT) avoid some unnecessary memory fragmentation... for composited objects that have a tr_completion, contain the it directly rather than a pointer to one allocated elsewhere on the heap.	2009-01-02 17:01:55 +00:00
Charles Kerr	d6b4db4a78	(trunk libT) add "inline" hint to several one-liner functions	2009-01-02 06:28:22 +00:00
Charles Kerr	e5fc032d9c	(trunk libT) Remove unused field: struct tr_peermsgs::peerSentBitfield	2009-01-01 18:35:21 +00:00
Charles Kerr	ad69a937ef	(trunk libT) have a pool of reusable evbuffers	2008-12-30 20:32:00 +00:00
Charles Kerr	53b25c7b49	(trunk libT) patch from wereHamster to simplify BT message parsing	2008-12-30 19:44:49 +00:00
Charles Kerr	56d04f76ad	(trunk libT) use a static buffer for fillOutputBuffer()	2008-12-30 08:25:39 +00:00
Charles Kerr	640c952ce8	(trunk libT) fix r7535 bug reported by Rolcol	2008-12-30 02:42:45 +00:00
Charles Kerr	f761311018	(trunk libT) avoid a few more unnecessary malloc() + free() pairs	2008-12-30 00:56:10 +00:00
Charles Kerr	c52d9fedab	(trunk libT) avoid some unnecessary memory fragmentation... for composited objects that have a tr_publisher, contain the it directly rather than a pointer to one allocated elsewhere on the heap.	2008-12-29 18:10:07 +00:00
Charles Kerr	dfcc9173e5	(trunk libT) #include "session.h" cleanup from wereHamster	2008-12-23 17:27:15 +00:00
Charles Kerr	acdd8c860c	(trunk libT) minor cleanups found while diffing for backport to 1.4x in r7455	2008-12-22 00:52:44 +00:00
Erick Turnquist	38efb62100	add some debugging stuff to track down where some bogus addresses are coming from	2008-12-21 19:13:52 +00:00
Charles Kerr	2b0ece7614	(trunk libT) remove tr_peermsgs.minActiveRequests based on wereHamster's feedback	2008-12-21 18:31:28 +00:00
Charles Kerr	fefa352b11	(trunk libT) remove unused PEER_PULSE_INTERVAL	2008-12-21 17:51:46 +00:00
Charles Kerr	a94e104040	(trunk libT) really fuck up the peer i/o code. also this breaks the mac build until someone removes iobuf.c from libtransmission's list of files.	2008-12-16 22:08:17 +00:00
Charles Kerr	9594112d27	updated email address	2008-12-16 00:20:44 +00:00
Charles Kerr	54be88dec3	(trunk libT) add ipv6 support by jhujhiti. I think this is the largest user-contributed patch we've ever used... thanks jhujhiti :)	2008-12-15 00:17:08 +00:00
Charles Kerr	0272708d32	(trunk libT) remove a couple of redundant and/or unused fields from struct peer_atom and tr_peermsgs.	2008-12-12 02:44:21 +00:00
Charles Kerr	451058d5cd	(libT) add some documentation about the three separate peer structs and how they are related.	2008-12-11 17:02:34 +00:00
Charles Kerr	65fe66c470	(trunk libT) fix bug which caused libtransmission to hold onto nonproductive peers for longer than it should've	2008-12-11 07:04:46 +00:00
Charles Kerr	718c1caa96	(trunk libT) #1565: give the peer requests a higher send priority. thanks to jusid for testing and sleuthing above & beyond the call of duty	2008-12-08 20:36:36 +00:00
Charles Kerr	81e636ad89	(libT) #1577: honor the peer's BEP 10 reqq key, if they provide it.	2008-12-08 20:23:10 +00:00
Charles Kerr	9e98dcd248	(libT) minor cleanup: using tr_bool, fixing up bad code indentation, using %zu on size_t instead of casting to int...	2008-12-05 22:56:19 +00:00
Charles Kerr	eb90a55608	(libT) #1562: 1.40 sends unnecessary `cancel' messages when choked	2008-12-05 01:12:55 +00:00
Charles Kerr	81c9bc0371	(libt) #1554: EPROTO (in peer-msgs.c) not defined on Windows	2008-12-03 13:21:41 +00:00
Charles Kerr	d4ee0d714a	(libT) #1379: support "partial seeds" BEP 22	2008-12-02 19:46:51 +00:00
Charles Kerr	341b40f192	libT) fix another r7234 bug reported by John_Clay	2008-12-02 19:38:46 +00:00
Charles Kerr	c414377c1a	(libT) fix r7234 bug reported by John_Clay	2008-12-02 19:06:08 +00:00
Charles Kerr	9d39ed2540	(libT) misc cleanup	2008-12-02 18:24:26 +00:00
Charles Kerr	f927ea5d63	(libT) #1549: support fast exensions' "reject" and "have all/none" messages	2008-12-02 17:10:54 +00:00
Charles Kerr	0d5fed6442	(libT) re-apply jhujhiti's IPv6 patch. This merges in my tr_port cleanup, so any new bugs are mine :/	2008-12-02 03:41:58 +00:00
Mitchell Livingston	c2b6fde6dd	#1548 Peers rapidly connect and reconnect after removing fast support.	2008-12-02 00:37:10 +00:00
Charles Kerr	7fa9813955	(libT) #252: revert r7195 (jhujhiti's IPv6 test patch). It seems to be the culprit in a `thrashing' issue reported by Waldorf and m1b in irc -- the behavior appeared between r7183 and r7187 -- so this commit is to trigger off a nightly build w/o the patch to test with in irc.	2008-12-01 20:21:06 +00:00
Charles Kerr	a588eaddc8	(libT) fix a small pex memory leak	2008-12-01 05:57:59 +00:00
Charles Kerr	37414ea274	(libT) fix memory leak introduced in r7186	2008-12-01 05:56:39 +00:00
Charles Kerr	f4f908be92	(libT) #1542: SIGSEGV in tr_publisherPublish	2008-11-30 21:36:49 +00:00
Charles Kerr	a2ad4f6543	(libT) patch from jhujhiti to add ipv6 support.	2008-11-30 00:47:18 +00:00
Charles Kerr	40b5fc5351	(libT) #1532: patch from jhujhiti to remove (unused) fast extensions	2008-11-29 20:37:34 +00:00
Charles Kerr	7fee78a697	(libT) re-enable lazy bitfields	2008-11-29 19:16:31 +00:00
Charles Kerr	d12af295c1	use tr_bool instead of C bitfields. (http://blogs.msdn.com/oldnewthing/archive/2008/11/26/9143050.aspx)	2008-11-28 22:11:41 +00:00
Charles Kerr	ca433d44ae	(libT) make peer-io's output buffer size more flexible based on the peer's speed	2008-11-28 16:00:29 +00:00
Charles Kerr	b46375fd8c	(libT) #1527: don't send pex messages if we don't have any new information to send	2008-11-28 05:48:17 +00:00
Charles Kerr	f44248a7d7	(libT) yet another stab at getting bandwidth management under control. this version may suck less than previous attempts. It also breaks the mac build until someone adds iobuf.[ch] to xcode...	2008-11-25 21:35:17 +00:00
Charles Kerr	be5e6774ad	(libT) #1468: another stab at getting the peer transfer speeds both fast and a little more consistent.	2008-11-24 04:21:23 +00:00
Charles Kerr	0bcff74d52	minor code cleanup	2008-11-23 16:30:09 +00:00
Charles Kerr	f172e8f96e	(libT) another possible fix for #1468: Speed display is very jumpy	2008-11-20 03:07:26 +00:00
Charles Kerr	a0fd42c5c3	(libT) better possible fix for #1468: Speed display is very jumpy	2008-11-17 04:00:57 +00:00
Charles Kerr	97e4efaa4d	(libT) #1474: lazy bitfields don't work quite right in 1.40	2008-11-16 08:56:18 +00:00