aboutsummaryrefslogtreecommitdiff
path: root/src/or/cpuworker.c
Commit message (Collapse)AuthorAge
* r9025@Kushana: nickm | 2006-09-29 18:33:13 -0400Nick Mathewson2006-09-29
| | | | | | | Differentiate more duplicated log entries svn:r8542
* r9023@Kushana: nickm | 2006-09-29 17:27:24 -0400Nick Mathewson2006-09-29
| | | | | | | Make distinct all non-bug messages at notice or higher that appear 3 or more times. svn:r8541
* r8725@Kushana: nickm | 2006-09-06 04:39:29 -0400Nick Mathewson2006-09-06
| | | | | | | spawn_func fixes: have cpuworker_main and dnsworker_main confirm to the right interfaces [casting func to void* is icky]. Also, make pthread_create() build without warnings. svn:r8327
* - made configure check if we are building for win32Mike Chiussi2006-09-06
| | | | | | | | | | | | | - made configure link to required system dll's if building for win32 - added diffs for libevent 1.1b - forced user to turn off eventdns if win32 is set - cleaned up tor_mmap_file()_win32 (not sure if it's stable) - cleaned up some warnings and typos svn:r8322
* r6908@Kushana: nickm | 2006-07-26 12:38:52 -0400Nick Mathewson2006-07-26
| | | | | | | Refactor connection_t into edge, or, dir, control, and base subtypes. This might save some RAM on busy exit servers, but really matters most in terms of correctness. svn:r6906
* Don't tell anybody, but we're going OO here. This patch splitsNick Mathewson2006-07-23
| | | | | | | | | | | | | | | circuit_t into origin_circuit_t and or_circuit_t. I fixed some segaults; there may be more. We still need to move more rendezvous stuff into subtypes. This is a trial run for splitting up connection_t; if the approach is insane, please say so soon so we can do something smarter. Also, this discards the old HALF_OPEN code, which nobody seems to want. svn:r6817
* Finally solve the "closing wedged cpuworkers" bug. Woo.Roger Dingledine2006-06-12
| | | | | | | | | | | This happened when we got two create cells in a row from the same TLS connection. It would hand one to the cpuworker, and then immediately handle the second one -- after it had registered that the first one was busy, but before it had updated the timestamp that we use to decide how *long* it's been busy. svn:r6606
* We got an obscure report of an assert error on a windows Tor server withRoger Dingledine2006-06-06
| | | | | | | | | | connection_add being called with socket = -1. The only places I can see where that happen would be if our tor_socketpair succeeds but it hands back negative values for some of its file descriptors. Perhaps this will produce more useful assert errors next time. svn:r6550
* Raise the timeout for complaining about wedged cpuworkers.Roger Dingledine2006-04-18
| | | | | | | | This value is high because some servers with low memory/cpu sometimes spend an hour or more swapping, and Tor starves. svn:r6406
* More cleanups noticed by weasel; also, remove macros that nobody uses.Nick Mathewson2006-03-12
| | | | svn:r6143
* Cleanup on time-relaqted constants. New conventions:Nick Mathewson2006-03-12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | 1) Surround all constants by (parens), whether we'll be using them in a denominator or not. 2) Express all time periods as products (24*60*60), not as multiplied-out constants (86400). 3) Comments like "(60*60) /* one hour */" are as pointless as comments like "c = a + b; /* set c to the sum of a and b */". Remove them. 4) All time periods should be #defined constants, not given inline. 5) All time periods should have doxygen comments. 6) All time periods, unless specified, are in seconds. It's not necessary to say so. To summarize, the old (lack of) style would allow: #define FOO_RETRY_INTERVAL 60*60 /* one hour (seconds) */ next_try = now + 3600; The new style is: /** How often do we reattempt foo? */ #define FOO_RETRY_INTERVAL (60*60) next_try = now + RETRY_INTERVAL; svn:r6142
* fix typo found in phobos's logsRoger Dingledine2006-02-15
| | | | svn:r6020
* more log conversions.Roger Dingledine2006-02-13
| | | | | | | whee. svn:r6003
* Happy new year!Roger Dingledine2006-02-09
| | | | svn:r5949
* Add reasons to DESTROY and RELAY_TRUNCATED cells.Nick Mathewson2006-01-05
| | | | svn:r5734
* disambiguate some debug entriesRoger Dingledine2005-12-26
| | | | svn:r5654
* Bite the bullet and limit all our source lines to 80 characters, the way IBM ↵Nick Mathewson2005-12-14
| | | | | | intended. svn:r5582
* answer nick's XXX, and muck with some log domain choicesRoger Dingledine2005-11-01
| | | | svn:r5339
* make circ->onionskin a pointer, not a static array. moria2 was usingRoger Dingledine2005-10-29
| | | | | | | | 125000 circuit_t's after it had been up for a few weeks, which translates to 20+ megs of wasted space. svn:r5333
* Per comments at the bottom of openssl/FAQ, call even more functions toNick Mathewson2005-10-25
| | | | | | | | clean up OpenSSL's toys when it's done playing. (Why isn't there an OpenSSL_free_everything() function?) svn:r5321
* Remove last vestiges of old logging interface.Nick Mathewson2005-10-25
| | | | svn:r5317
* Change more files to new loggin interface. 3 left.Nick Mathewson2005-10-25
| | | | svn:r5310
* Downgrade a few INFO level logs to DEBUG again. Also add two or three newPeter Palfrader2005-10-17
| | | | | | | | logs in cases where a calling function's log was downgraded and we wouldn't get any log message otherwise. svn:r5263
* there was no need to die when we failed to spawn a cpuworker.Roger Dingledine2005-09-23
| | | | | | | it handles it fine if we decide not to die. svn:r5114
* rotate dns and cpu workers if the controller changes options thatRoger Dingledine2005-08-15
| | | | | | | will affect them. svn:r4787
* initialize the dns worker cache tree whether or not we startRoger Dingledine2005-08-15
| | | | | | | out as a server. svn:r4785
* Make errors retrievable from tor_socketpair; resolve bug 163.Nick Mathewson2005-06-30
| | | | svn:r4509
* Docment or add DOCDOC comments to undocumented functions in src/or. Make ↵Nick Mathewson2005-06-11
| | | | | | function definition format uniform. svn:r4411
* flesh out the source file descriptions for doxygenRoger Dingledine2005-06-11
| | | | svn:r4404
* Change end-of-file NLNL convention. It turns out arma I and I agree.Nick Mathewson2005-06-09
| | | | svn:r4382
* New whitespace normalization rule: no blank line at EOF.Nick Mathewson2005-06-09
| | | | svn:r4378
* Make Tor compile with no warnings with gcc4.0 on OSXNick Mathewson2005-05-07
| | | | svn:r4184
* Call tor_free_all instead of connections_free_all after forkingNick Mathewson2005-05-03
| | | | svn:r4173
* New and frightening code to implement fast-path first-hop CREATE_FAST cells. ↵Nick Mathewson2005-05-02
| | | | | | Watch out when we bump the version to 0.1.0.6-rc! svn:r4162
* only complain about wedged cpuworkers after an hour, not 100 secsRoger Dingledine2005-04-23
| | | | svn:r4100
* If it has not been 100 seconds since we told the cpuworker about an onion, ↵Nick Mathewson2005-04-08
| | | | | | complain! Oh, wait... svn:r4059
* add function commentRoger Dingledine2005-04-07
| | | | svn:r4043
* We have a bug that I can't find. Sometimes, very rarely, cpuworkers getRoger Dingledine2005-04-07
| | | | | | | | | stuck in the 'busy' state, even though the cpuworker process thinks of itself as idle. I don't know why. But here's a workaround to kill any cpuworker that's been busy for more than 100 seconds. svn:r4042
* fix an assert trigger: when we have the rare case of acceptingRoger Dingledine2005-04-06
| | | | | | | | a conn on 0.0.0.0:0, then when we look through the connection array, we'll find any of the workers. this is no good. svn:r4027
* Hopefully, this will make ORs much faster, and not break them: keep a big ↵Nick Mathewson2005-04-06
| | | | | | splay tree of (circid,orconn)->circuit mappings to make circuit_get_by_circid_conn much faster. svn:r4020
* Make last fix work on windows too.Nick Mathewson2005-04-04
| | | | svn:r4009
* Fix possible bug on threading platforms with running out of fds do to ↵Nick Mathewson2005-04-04
| | | | | | cpuworkers and dnsworkers. svn:r4008
* update copyright notices.Nick Mathewson2005-04-01
| | | | svn:r3982
* Change from inet_ntoa to a threadproof tor_inet_ntoa.Nick Mathewson2005-02-22
| | | | svn:r3656
* forward-port the shutting upRoger Dingledine2005-01-28
| | | | svn:r3450
* Handle unavailable hidden services better. We try each intro pointRoger Dingledine2005-01-19
| | | | | | | | | | until none are left, then we try to refetch the descriptor. If it's the same one we had before, then close streams right then. Whenever a new stream arrives, even if it's right after, optimistically try refetching the descriptor, just in case. svn:r3379
* Give more useful log messages when the tor process closes dns/cpuworker ↵Nick Mathewson2005-01-19
| | | | | | connections. "Error: Success!" makes us seem like crazy people." svn:r3373
* More work on task #43: fix race conditions on multithreaded (win32) servers.Nick Mathewson2005-01-03
| | | | svn:r3251
* Possible fix for task #43: when running on a multithreaded environment ↵Nick Mathewson2005-01-03
| | | | | | (currently only windows), threads should not close opposite sides of their socketpairs, and workers should not call connection_free_all(). This may fix win32 servers. svn:r3247
* More win32 fixes: 1) tolerate extra "readable" events better. 2) when being ↵Nick Mathewson2004-12-06
| | | | | | multithreaded, leave parent fdarray open. svn:r3092
generated by cgit v1.2.3 (git 2.25.4) at 2024-11-06 05:49:49 +0000