aboutsummaryrefslogtreecommitdiff
path: root/src
Commit message (Expand)AuthorAge
* r19079@catbus: nickm | 2008-03-26 17:16:46 -0400••• Free authority certs on exit. Backport candidate. svn:r14205 Nick Mathewson2008-03-26
* r19074@catbus: nickm | 2008-03-26 17:08:32 -0400••• Start new address policies with refcount of 1, not 2. Backport candidate once tested more. svn:r14204 Nick Mathewson2008-03-26
* fix a leak when adding a temp log•••svn:r14203 Roger Dingledine2008-03-26
* fix mem leak in parsing log config lines•••svn:r14202 Roger Dingledine2008-03-26
* r19072@catbus: nickm | 2008-03-26 13:50:24 -0400••• Add code to debug memory area size. Use results of this code to set a couple of area sizes more sanely. svn:r14201 Nick Mathewson2008-03-26
* r19066@catbus: nickm | 2008-03-26 13:24:15 -0400••• Initialize logging before we log anything. Bugfix on 0.2.0.x-alpha. svn:r14199 Nick Mathewson2008-03-26
* r19062@catbus: nickm | 2008-03-26 12:56:25 -0400••• Fix whitespace svn:r14197 Nick Mathewson2008-03-26
* r19061@catbus: nickm | 2008-03-26 12:53:18 -0400••• Now that every thing in routerparse.c is switched over to use memareas, there is no need to keep the heap-allocated token code. svn:r14196 Nick Mathewson2008-03-26
* r19060@catbus: nickm | 2008-03-26 12:44:19 -0400••• Make v2 hidden service descriptors use the new area allocation logic. This works for me, but Karsten should definitely have a look at it. svn:r14195 Nick Mathewson2008-03-26
* r19049@catbus: nickm | 2008-03-26 12:33:25 -0400••• Add new stacklike, free-all-at-once memory allocation strategy. Use it when parsing directory information. This helps parsing speed, and may well help fragmentation some too. hidden-service-related stuff still uses the old tokenizing strategies. svn:r14194 Nick Mathewson2008-03-26
* r19041@catbus: nickm | 2008-03-25 16:20:42 -0400••• More unit tests to improve coverage. svn:r14185 Nick Mathewson2008-03-25
* r19039@catbus: nickm | 2008-03-25 12:15:58 -0400••• Add some unit tests, particularly for AES counter mode. svn:r14180 Nick Mathewson2008-03-25
* Mar 24 08:43:30.014 [debug] circuit_find_to_cannibalize(): Hunting for•••a circ to cannibalize: purpose 5, uptime 0, capacity 4, internal 0 svn:r14166 Roger Dingledine2008-03-24
* correct a confusing log message•••svn:r14165 Roger Dingledine2008-03-24
* fix r14162 so it was what i meant. sometimes precedence is surprising!•••svn:r14164 Roger Dingledine2008-03-24
* when our onehop circuits were failing, nobody was telling the•••streams. so they waited 120 seconds before timing out. this was particularly bad during bootstrapping, if an authority is down or not answering right. svn:r14163 Roger Dingledine2008-03-24
* If our create-fast cell is mysteriously never answered for a begindir•••attempt, notice more quickly. Some of our bootstrapping attempts have a 60 second delay while we sit there wondering why we're getting no response. svn:r14162 Roger Dingledine2008-03-24
* and one more•••svn:r14152 Roger Dingledine2008-03-21
* make it build, too•••svn:r14151 Roger Dingledine2008-03-21
* r19004@catbus: nickm | 2008-03-21 15:18:43 -0400••• Use RAND_poll() again: the bug that made us stop using it has been fixed. svn:r14150 Nick Mathewson2008-03-21
* forward-port 14113-r14115•••svn:r14116 Roger Dingledine2008-03-18
* r18941@catbus: nickm | 2008-03-18 14:59:43 -0400••• Likely fix for bug 632: do not call start_reading on a placeholder connection. svn:r14109 Nick Mathewson2008-03-18
* r18937@catbus: nickm | 2008-03-18 14:50:39 -0400••• Fix the other lingering part of bug 617: make ClientDNSRejectInternalAddresses actually work. svn:r14107 Nick Mathewson2008-03-18
* r18929@catbus: nickm | 2008-03-18 12:08:16 -0400••• Detect errors from directory listing correctly on win32. Bug found by lodger. svn:r14102 Nick Mathewson2008-03-18
* r18927@catbus: nickm | 2008-03-18 11:11:49 -0400••• Combine common code in set_max_file_descriptors(): all that varies from platform to platform in the no-getrlimit() case is the connection limit and the platform name. svn:r14101 Nick Mathewson2008-03-18
* r18923@catbus: nickm | 2008-03-18 11:01:22 -0400••• Add missing typecasts to log message in set_max_file_descriptors to tell gcc everything is okay on windows. Fixes bug 630. svn:r14099 Nick Mathewson2008-03-18
* r18919@catbus: nickm | 2008-03-18 10:53:38 -0400••• Forward-port bug 622 fix as diagnosed by lodger. svn:r14096 Nick Mathewson2008-03-18
* r18913@catbus: nickm | 2008-03-18 10:30:39 -0400••• 16, not 64. svn:r14094 Nick Mathewson2008-03-18
* point out another bug for nick. and if it's *not* a bug, that's•••stunning and i want to know why. :) svn:r14088 Roger Dingledine2008-03-18
* forward-port the 0.2.0.22-rc changelog entry, plus remove•••some duplicate items. svn:r14086 Roger Dingledine2008-03-18
* r18900@catbus: nickm | 2008-03-17 18:32:25 -0400••• Fix tests and routerdesc behavior on trunk. svn:r14084 Nick Mathewson2008-03-17
* r18896@catbus: nickm | 2008-03-17 16:10:54 -0400••• Fix bug in earlier bugfix. Note stupidness of allowing NULL policies at all. Disallow empty exit policies in router descriptors. svn:r14082 Nick Mathewson2008-03-17
* r18891@catbus: nickm | 2008-03-17 13:20:20 -0400••• Patch from Sebastian Hahn: give an error message on "error replacing old router store." svn:r14080 Nick Mathewson2008-03-17
* r18890@catbus: nickm | 2008-03-17 13:19:29 -0400••• Clarify documentation for file_status a little svn:r14079 Nick Mathewson2008-03-17
* r18880@catbus: nickm | 2008-03-17 12:51:24 -0400••• Fix policy-related crash bug found by lodger. svn:r14077 Nick Mathewson2008-03-17
* And use 16k pages on ia64. Maybe this should be a configure thing•••svn:r14073 Peter Palfrader2008-03-17
* r18878@catbus: nickm | 2008-03-17 00:11:02 -0400••• Clean up an overwide line. svn:r14072 Nick Mathewson2008-03-17
* r18873@catbus: nickm | 2008-03-17 00:06:31 -0400••• oops. guard status logging change had a bug. it is non-obvious how to make the code perfect atm; just make it work. svn:r14070 Nick Mathewson2008-03-17
* r18872@catbus: nickm | 2008-03-16 23:56:48 -0400••• Only dump all guard node status to the log when the guard node status actually changes. Downgrade the 4 most common remaining INFO log messages to DEBUG. svn:r14069 Nick Mathewson2008-03-17
* r18862@catbus: nickm | 2008-03-16 23:33:11 -0400••• Part of fix for bug 617: allow connection_ap_handshake_attach_circuit() to mark connections, to avoid double-mark warnings. Note that this is an incomplete refactoring. svn:r14066 Nick Mathewson2008-03-17
* r18861@catbus: nickm | 2008-03-16 23:22:56 -0400••• Fix a couple of bugs in setting control log callback severity. svn:r14065 Nick Mathewson2008-03-17
* r18852@catbus: nickm | 2008-03-16 22:47:19 -0400••• Downgrade "sslv3 alert handshake failure" message to info. svn:r14057 Nick Mathewson2008-03-17
* r18850@catbus: nickm | 2008-03-16 22:35:38 -0400••• Use 8k pages in openbsd malloc code on alpha. Bug and solution found by weasel. Also, when initializing openbsd malloc code, check that compiled page size matches output of getpagesize(). svn:r14056 Nick Mathewson2008-03-17
* r18804@catbus: nickm | 2008-03-13 18:18:31 -0400••• Refactor log domain mask code so that nobody outside of log.c has to use SEVERITY_MASK_IDX. It is error-prone. svn:r14016 Nick Mathewson2008-03-13
* r18803@catbus: nickm | 2008-03-13 17:59:25 -0400••• Fix behavior of switch_logs_debug() in trunk. Fixes bug 626. Bugfix on r13875. svn:r14015 Nick Mathewson2008-03-13
* r18793@catbus: nickm | 2008-03-13 14:09:19 -0400••• Add a malloc_good_size() implementation to OpenBSD_malloc_Linux.c. Also, make configure.in not use support functions for the platform malloc when we are not using the platform mallocs. svn:r14010 Nick Mathewson2008-03-13
* Assert that severity in logv() is sane. Interesting effects otherwise (It is...•••svn:r14006 Peter Palfrader2008-03-13
* downgrade get_interface_address6 logging severity to LOG_DEBUG.•••get_interface_address6() fails regardless of the allocator used, wever logging to the original severity of 0 causes an assert error only with the bsd allocator. weird. svn:r14005 Peter Palfrader2008-03-13
* r18787@catbus: nickm | 2008-03-13 11:11:52 -0400••• Make set-option functions return sensible error codes from an enum, not mysterious negative integers svn:r14004 Nick Mathewson2008-03-13
* Use proper log levels with get_interface_address6() calls so we do not die in...•••svn:r14003 Peter Palfrader2008-03-13
generated by cgit v1.2.3 (git 2.47.1) at 2025-03-14 01:26:37 +0000