Commit Graph

18231 Commits

Author SHA1 Message Date
Nick Mathewson
f425cf8338 Start writing tests for 10169.
Now we cover more chunk allocation functions.
2014-02-12 12:47:49 -05:00
Nick Mathewson
eb6f433bdb Debugging code inbuffers.c for debugging chunk allocation.
Currently on; will disable later in this branch.
2014-02-12 12:46:17 -05:00
Nick Mathewson
87fb1e324c Merge remote-tracking branch 'public/bug10169_024' into bug10169_025_v2
Conflicts:
	src/or/circuitlist.c
2014-02-12 12:44:58 -05:00
Nick Mathewson
05d8111eed Merge remote-tracking branch 'public/bug10169_023' into bug10169_024 2014-02-12 12:39:15 -05:00
Nick Mathewson
fd28754dd3 Actually release buffer freelists when handling OOM conditions.
Otherwise freeing buffers won't help for a little while.
2014-02-12 12:38:20 -05:00
Nick Mathewson
7951591744 Fix bugs in bug10169 bugfix memory tracking
The chunk_grow() and chunk_copy() functions weren't adjusting the
memory totals properly.

Bugfix not on any released Tor version.
2014-02-12 12:37:41 -05:00
Nick Mathewson
10d1b0b33e Changes file for feature4900 2014-02-12 12:20:29 -05:00
Nick Mathewson
c1e98c8afe Randomize the global siphash key at startup
This completes our conversion to using siphash for our hash functions.
2014-02-12 12:12:58 -05:00
Nick Mathewson
d3fb846d8c Split crypto_global_init() into pre/post config
It's increasingly apparent that we want to make sure we initialize our
PRNG nice and early, or else OpenSSL will do it for us.  (OpenSSL
doesn't do _too_ bad a job, but it's nice to do it ourselves.)

We'll also need this for making sure we initialize the siphash key
before we do any hashes.
2014-02-12 12:04:07 -05:00
Nick Mathewson
0e97c8e23e Siphash-2-4 is now our hash in nearly all cases.
I've made an exception for cases where I'm sure that users can't
influence the inputs.  This is likely to cause a slowdown somewhere,
but it's safer to siphash everything and *then* look for cases to
optimize.

This patch doesn't actually get us any _benefit_ from siphash yet,
since we don't really randomize the key at any point.
2014-02-12 11:32:10 -05:00
Nick Mathewson
f05820531a csiphash: Add functions to take a global key. 2014-02-12 11:27:41 -05:00
Nick Mathewson
f51df9bb93 Tests for siphash, from reference implementation. 2014-02-12 10:58:03 -05:00
Nick Mathewson
1bd86b69b6 csiphash: avoid unaligned access on non-x86 2014-02-12 10:57:46 -05:00
Nick Mathewson
9605978eb6 Get csiphash better integrated with our build system 2014-02-12 10:24:04 -05:00
Nick Mathewson
f4656c0cc9 Raw import of Marek Majkowski's cisphash.c
siphash is a hash function designed for producing hard-to-predict
64-bit outputs from short inputs and a 128-bit key.  It's chosen for
security and speed.

See https://131002.net/siphash/ for more information on siphash.

Source: https://github.com/majek/csiphash/
2014-02-12 10:09:45 -05:00
Nick Mathewson
2c0088b8aa Merge remote-tracking branch 'origin/maint-0.2.4' 2014-02-12 09:17:11 -05:00
Nick Mathewson
d6e6eaba60 Fix windows compilation of e0c8031516
There is no WSAEPERM; we were implying that there was.This fixes a
bug in e0c8031516, which hadn't yet
appeared in any released Tor.
2014-02-12 09:16:22 -05:00
Karsten Loesing
74c2bff781 Remove remaining v0 hidden service descriptor code.
Fixes the rest of #10841 after #10881 already removed some hidden service
authority code.
2014-02-12 14:36:08 +01:00
Roger Dingledine
1ccc282122 add a missing word to the man page 2014-02-12 04:01:59 -05:00
Roger Dingledine
7c39bf0e44 go through and rewrite the changes files to be more user-facing 2014-02-12 03:59:59 -05:00
Nick Mathewson
0ee449ca92 Merge remote-tracking branch 'origin/maint-0.2.4' 2014-02-11 18:58:58 -05:00
Nick Mathewson
91d4bb0b00 Merge branch 'bug10777_netunreach_024' into maint-0.2.4 2014-02-11 18:57:55 -05:00
Nick Mathewson
0844a262a2 update changes file 2014-02-11 18:57:38 -05:00
Nick Mathewson
8836c1ee2f Merge remote-tracking branch 'public/bug10777_nointernal_024' into maint-0.2.4 2014-02-11 18:55:26 -05:00
Nick Mathewson
c3720c05fa Free leakable values in the unit tests.
Thanks, Coverity!  (CID 1171414, 1171415, 1171416)
2014-02-11 18:10:33 -05:00
Nick Mathewson
10d4d3e2d5 Merge remote-tracking branch 'public/no_itime_queue_025' 2014-02-11 11:52:35 -05:00
Nick Mathewson
5e0cdc5ef2 Merge branch 'bug10881' 2014-02-11 11:42:06 -05:00
Nick Mathewson
0228b9bd4b Tweak DirAuthority documentation
Per a comment from karsten, there is no longer v2 authority support,
so stop pretending there is. Also, fix a grammar error.x
2014-02-11 11:34:32 -05:00
Nick Mathewson
cb065a55bd Merge remote-tracking branch 'karsten/one-dirauth' 2014-02-11 10:15:03 -05:00
Nick Mathewson
cb28fe486f Merge remote-tracking branch 'public/bug10722' 2014-02-11 10:08:58 -05:00
Karsten Loesing
aae4ebf63f Add changes file for ticket 10842.
This is a bugfix on 0.2.2.26-beta, because 6b83b3b made directory
authorities remove themselves from the list of directory authorities to
upload to, but didn't suppress the warning in case they're the only
directory authority in the network.
2014-02-11 08:51:28 +01:00
Nick Mathewson
c0483c7f85 Remove options for configuring HS authorities.
(There is no longer meaningfully any such thing as a HS authority,
since we stopped uploading or downloading v0 hs descriptors in
0.2.2.1-alpha.)

Implements #10881, and part of #10841.
2014-02-10 22:41:52 -05:00
Roger Dingledine
dd3f2f6332 fix trivial typo 2014-02-10 16:05:35 -05:00
Nick Mathewson
e0c8031516 make EACCES survivable too. 2014-02-10 15:06:10 -05:00
Nick Mathewson
5b55e0e181 Merge remote-tracking branch 'public/no_itime_queue'
The conflicts here were tricky, and required me to eliminate the
command-queue as well.  That wasn't so hard.

Conflicts:
	src/or/or.h
	src/or/relay.c
2014-02-10 15:04:23 -05:00
Nick Mathewson
3133cde3c1 Excise the insertion_time_elem_t logic
It's now redundant with the inserted_time field in packed_cell_t

Fixes bug 10870.
2014-02-10 13:55:27 -05:00
Nick Mathewson
a73b0da653 Merge remote-tracking branch 'origin/maint-0.2.4' 2014-02-09 21:41:59 -05:00
Nick Mathewson
7f6aa780e3 Merge remote-tracking branch 'andrea/bug9602' into maint-0.2.4 2014-02-09 21:41:24 -05:00
Nick Mathewson
b15f75b632 Don't treat END_STREAM_REASON_INTERNAL as total circuit failure
It can happen because we sent something that got an ENETUNREACH
response.

Bugfix on 0.2.4.8-alpha; fixes a part of bug 10777.
2014-02-09 21:35:14 -05:00
Nick Mathewson
f5d32c08ba Call ENETUNREACH a case of NOROUTE, not a case of INTERNAL.
Found by cypherpunks; fix for a part of bug 10777; bugfix on 0.1.0.1-rc.
2014-02-09 21:30:23 -05:00
Andrea Shepard
c330d63ff7 Make sure orconn->chan gets nulled out when channels exit from channel_free_all() too 2014-02-08 14:05:51 -08:00
Karsten Loesing
26dd328891 Update to the February 2014 GeoIP database. 2014-02-08 12:09:37 +01:00
Karsten Loesing
749ead79ad Don't zero-pad day of the month. 2014-02-08 12:03:27 +01:00
Karsten Loesing
3bea0fff79 Add script to convert GeoLite2 country databases.
Copied over from https://github.com/nmathewson/mmdb-convert,
commit 52f0d43408ed3a0fe7c3b2fa7099da7f6b7c4502.
2014-02-08 11:56:54 +01:00
Karsten Loesing
becc3d5b4e Remove script to clean up GeoLite country databases. 2014-02-08 11:56:27 +01:00
Karsten Loesing
ebe7e22045 Suppress warning in networks with only 1 dirauth. 2014-02-08 11:02:27 +01:00
Nick Mathewson
a0577aacb4 Merge remote-tracking branch 'origin/maint-0.2.4' 2014-02-07 17:41:41 -05:00
Nick Mathewson
9bb34aa897 Survive fedora's openssl in our benchmarks
Apparently fedora currently has ECDH but not P224. This isn't a huge
deal, since we no longer use OpenSSL's P224 ever (see #9780 and
72c1e5acfe). But we shouldn't have segfaulting benchmarks really.

Fixes bug 10835; bugfix on 0.2.4.8-alpha.
2014-02-07 17:36:11 -05:00
Nick Mathewson
fdf68479b0 Explain CURVE25519_ENABLED: closes 9774 2014-02-07 16:16:56 -05:00
Nick Mathewson
aa3c8c1397 Use the right functions; strncpy is usually not the answer 2014-02-07 12:25:49 -05:00