summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)Author
2009-01-03Another attempt at fixing coverity scan CID 361.Nick Mathewson
svn:r17863
2009-01-02Only set sin_len/sin6_len when they exist.Nick Mathewson
svn:r17851
2009-01-02take out my IMPOSSIBLE_TO_DOWNLOAD+1 hackRoger Dingledine
svn:r17850
2009-01-02Try harder to make sure we zero-out the extraneous sockaddr fields and that ↵Nick Mathewson
we set sockaddr_len. Conceivably a backport candidate, though nothing has yet been sen to break. svn:r17849
2008-12-31Increment n_download_failures up to IMPOSSIBLE_TO_DOWNLOAD-1, not up to ↵Nick Mathewson
IMPOSSIBLE_TO_DOWNLOAD. svn:r17835
2008-12-31switch over the bridge descriptor download mechanism toRoger Dingledine
use the same download mechanism as other places. i had to make an ugly hack around "IMPOSSIBLE_TO_DOWNLOAD+1". we should unhack that sometime. svn:r17834
2008-12-31fix a bug in download failure logging; don't do this "stopRoger Dingledine
downloading forever" thing with bridge descriptor downloads. svn:r17833
2008-12-30Debug code for task 878.Karsten Loesing
svn:r17828
2008-12-30Remove some code that is #ifdefed out, and that we no longer seem to use, if ↵Nick Mathewson
we ever did. svn:r17827
2008-12-30Indeed, arma was right. There was one ntohl too many in debug_ntop in ↵Nick Mathewson
eventdns.c svn:r17826
2008-12-30point out the bug location to nick. unless i'm wrong.Roger Dingledine
svn:r17825
2008-12-29Downgrade some xxx021s, comment more on others, etcNick Mathewson
svn:r17823
2008-12-29Actually log reasons on unusable guards again.Nick Mathewson
svn:r17822
2008-12-29Spec compliance: if weighted MTBF of a router is at least 5 days, always ↵Nick Mathewson
vote it as Stable. svn:r17821
2008-12-29Refactor sockaddr family match check into a new functionNick Mathewson
svn:r17820
2008-12-29Explain why the XXXX021 in connection_or.c was there, and explain why it ↵Nick Mathewson
doesn' actually need to get fixed in 0.2.1, I think. svn:r17819
2008-12-29Replace kludgy assert with something a little less awful, and avoid needless ↵Nick Mathewson
strlcpy in rend descriptor encoding. svn:r17818
2008-12-29Downgrade the last xxx021 in routerparse. The duplicate code stands for now.Nick Mathewson
svn:r17817
2008-12-29Do not not cannibalize a circuit that has run out of RELAY_EARLY cells. ↵Nick Mathewson
Partial bug 878 fix. svn:r17815
2008-12-29Do not cannibalize a circuit that has run out of RELAY_EARLY cells. Partial ↵Nick Mathewson
bug 878 fix. svn:r17814
2008-12-29Extact parse-the-token-arguments to its own function, and change it to a ↵Nick Mathewson
single-pass algorithm. This simplifies the parsing code and speeds it up a little. svn:r17812
2008-12-29Remove a call to find_whitespace_eos that didn't actually do anything.Nick Mathewson
svn:r17811
2008-12-29While I'm at it, refactor control.c a little so that the dead code no longer ↵Nick Mathewson
exists. svn:r17810
2008-12-29coverity thinks it's dumb to check networkstatus_v2_list in one place and ↵Nick Mathewson
not another. Coverity has a point, even though the check may be redundant. CID 361. svn:r17809
2008-12-29More compilation tweaks on Android: fix two clear errors in our code that ↵Nick Mathewson
apparently the compiler cares about there. svn:r17808
2008-12-29Use a consistent naming standard for header file guard macros, taking care ↵Nick Mathewson
not to collide with any system headers. This tripped us up on Android. svn:r17805
2008-12-29Use RSA_generate_key_ex where available.Nick Mathewson
svn:r17804
2008-12-29Switch address comparisons in policies to be exact rather than semantic. ↵Nick Mathewson
Until we do ipv6 exit policies and until we know whether we even allow ::ffff:0:0/96 addresses, there is no point in doing "semantic" comparisons. This was also showing up on oprofile. svn:r17803
2008-12-29Refactor some exit-policy-related functions that showed up in oprofile.Nick Mathewson
Specifically, split compare_tor_addr_to_addr_policy() from a loop with a bunch of complicated ifs inside into some ifs, each with a simple loop. Rearrange router_find_exact_exit_enclave() to run a little faster. Bizarrely, router_policy_rejects_all() shows up on oprofile, so precalculate it per routerinfo. svn:r17802
2008-12-29Fix bug in recent address.c patch: actually set the value of address * to ↵Nick Mathewson
0.0.0.0 as we did before. This makes CMP_EXACT comparisons with bitmask 0 work on address * again. svn:r17801
2008-12-27Fix compilation under gethostbyname-based systems.Nick Mathewson
svn:r17800
2008-12-27remove the responsibility for setting listensocklen to the function that ↵Nick Mathewson
made the sockaddr svn:r17799
2008-12-27partial move to letting bridge descriptor fetches use our new (well,Roger Dingledine
new from their perspective) directory download schedule abstraction. not done yet, but i'd better get this out of my sandbox before nick does another sweeping change. :) svn:r17798
2008-12-27New controller event "clients_seen" to report a geoip-based summaryRoger Dingledine
of which countries we've seen clients from recently. Now controllers like Vidalia can show bridge operators that they're actually making a difference. svn:r17796
2008-12-27Remove a deprecated controller alias; make another one generate a warning.Nick Mathewson
svn:r17793
2008-12-27Make connection_create_listener take a sockaddr_len, not just a sockaddr ↵Nick Mathewson
whose length we try to divine. svn:r17792
2008-12-26Add some early checks to keep ipv6 addresses from failing by accident. Now, ↵Nick Mathewson
they fail on purpose, at least till 0.2.2.something. svn:r17791
2008-12-26Refactor tor_addr_t manipulation functions so that as few as possible look ↵Nick Mathewson
at the tor_addr_t representation. svn:r17790
2008-12-26get_interface_addr6(), and by extension get_interface_addr(), were pretty ↵Nick Mathewson
borked. Copying a tor_addr_t from a sockaddr_storage using memcpy is a poor notion. svn:r17789
2008-12-26Refactor tor_addr_compare_masked() so that CMP_SEMANTIC makes more sense, ↵Nick Mathewson
and has decent semantics for maskbits; and so that CMP_EXACT works right for bits==0. svn:r17788
2008-12-26Bump version to 0.2.1.9-alpha-devNick Mathewson
svn:r17787
2008-12-26Document and make more conservative the logic for picking the size of the ↵Nick Mathewson
bloom filter to use for retaining descriptors. svn:r17786
2008-12-26Document our Bloom filter parameter choices.Nick Mathewson
svn:r17785
2008-12-26Document the pk operation count fields in rephist.c, and move them into a ↵Nick Mathewson
struct for clarity, and change their type to what we actually want to cast them to. svn:r17784
2008-12-26Add more missing documentation, and correct an error in container.c ↵Nick Mathewson
documentation: Don't introduce two parameters called n when you're calling an algorithm O(n). svn:r17783
2008-12-25prepare for 0.2.1.9-alphaRoger Dingledine
svn:r17778
2008-12-25ok, fair enough, some of these variables were time_t's after allRoger Dingledine
svn:r17777
2008-12-25so which is it? TIME_MAX or INT_MAX? pick INT_MAX for now.Roger Dingledine
svn:r17772
2008-12-25some fixes i found in my sandboxRoger Dingledine
svn:r17771
2008-12-24Use the literal parse of an address in dns_resolve_impl if parsing the ↵Nick Mathewson
address as an ipv4 or ipv6 address _succeeded_. Not if it failed. Bug introduced in r17707 (post 0.2.1.8-alpha), and found by xiando. svn:r17758
7apu2bq3rhoylwmucxizk54afw5jyda7pho4j5zdcqpwkufke7zdzwad.onion