]> git.ipfire.org Git - thirdparty/tor.git/log
thirdparty/tor.git
11 years agoforward-port recent changelogs and release notes
Roger Dingledine [Sun, 22 Dec 2013 09:30:06 +0000 (04:30 -0500)] 
forward-port recent changelogs and release notes

11 years agoMerge remote-tracking branch 'origin/maint-0.2.4'
Nick Mathewson [Sun, 22 Dec 2013 02:16:06 +0000 (21:16 -0500)] 
Merge remote-tracking branch 'origin/maint-0.2.4'

11 years agoMerge branch 'bug10456' into maint-0.2.4
Nick Mathewson [Sun, 22 Dec 2013 02:15:39 +0000 (21:15 -0500)] 
Merge branch 'bug10456' into maint-0.2.4

11 years agoFix a logic error in circuit_stream_is_being_handled.
Nick Mathewson [Sat, 21 Dec 2013 15:15:09 +0000 (10:15 -0500)] 
Fix a logic error in circuit_stream_is_being_handled.

When I introduced the unusable_for_new_circuits flag in
62fb209d837f3f551, I had a spurious ! in the
circuit_stream_is_being_handled loop.  This made us decide that
non-unusable circuits (that is, usable ones) were the ones to avoid,
and caused it to launch a bunch of extra circuits.

Fixes bug 10456; bugfix on 0.2.4.12-alpha.

11 years agoMerge remote-tracking branch 'linus/bug10324'
Nick Mathewson [Fri, 20 Dec 2013 16:40:21 +0000 (11:40 -0500)] 
Merge remote-tracking branch 'linus/bug10324'

11 years agoMerge remote-tracking branch 'origin/maint-0.2.4'
Nick Mathewson [Thu, 19 Dec 2013 03:04:21 +0000 (22:04 -0500)] 
Merge remote-tracking branch 'origin/maint-0.2.4'

Conflicts:
src/common/crypto.c

11 years agoMerge branch 'bug10402_redux_024' into maint-0.2.4
Nick Mathewson [Thu, 19 Dec 2013 02:55:06 +0000 (21:55 -0500)] 
Merge branch 'bug10402_redux_024' into maint-0.2.4

11 years agoNever allow OpenSSL engines to replace the RAND_SSLeay method
Nick Mathewson [Wed, 18 Dec 2013 16:49:44 +0000 (11:49 -0500)] 
Never allow OpenSSL engines to replace the RAND_SSLeay method

This fixes bug 10402, where the rdrand engine would use the rdrand
instruction, not as an additional entropy source, but as a replacement
for the entire userspace PRNG.  That's obviously stupid: even if you
don't think that RDRAND is a likely security risk, the right response
to an alleged new alleged entropy source is never to throw away all
previously used entropy sources.

Thanks to coderman and rl1987 for diagnosing and tracking this down.

11 years agoMerge remote-tracking branch 'origin/maint-0.2.4'
Nick Mathewson [Tue, 17 Dec 2013 18:54:02 +0000 (13:54 -0500)] 
Merge remote-tracking branch 'origin/maint-0.2.4'

11 years agoMerge branch 'bug10423' into maint-0.2.4
Nick Mathewson [Tue, 17 Dec 2013 18:53:11 +0000 (13:53 -0500)] 
Merge branch 'bug10423' into maint-0.2.4

11 years agoMerge remote-tracking branch 'origin/maint-0.2.4'
Nick Mathewson [Tue, 17 Dec 2013 18:18:00 +0000 (13:18 -0500)] 
Merge remote-tracking branch 'origin/maint-0.2.4'

Conflicts:
src/or/microdesc.c

Conflict because one change was on line adjacent to line where
01206893 got fixed.

11 years agoMerge remote-tracking branch 'public/bug10409_023' into maint-0.2.4
Nick Mathewson [Tue, 17 Dec 2013 18:15:45 +0000 (13:15 -0500)] 
Merge remote-tracking branch 'public/bug10409_023' into maint-0.2.4

11 years agoAvoid double-free on failure to dump_descriptor() a cached md
Nick Mathewson [Tue, 17 Dec 2013 18:12:52 +0000 (13:12 -0500)] 
Avoid double-free on failure to dump_descriptor() a cached md

This is a fix for 10423, which was introducd in caa0d15c in 0.2.4.13-alpha.

Spotted by bobnomnom.

11 years agoAvoid free()ing from an mmap on corrupted microdesc cache
Nick Mathewson [Mon, 16 Dec 2013 18:00:15 +0000 (13:00 -0500)] 
Avoid free()ing from an mmap on corrupted microdesc cache

The 'body' field of a microdesc_t holds a strdup()'d value if the
microdesc's saved_location field is SAVED_IN_JOURNAL or
SAVED_NOWHERE, and holds a pointer to the middle of an mmap if the
microdesc is SAVED_IN_CACHE.  But we weren't setting that field
until a while after we parsed the microdescriptor, which left an
interval where microdesc_free() would try to free() the middle of
the mmap().

This patch also includes a regression test.

This is a fix for #10409; bugfix on 0.2.2.6-alpha.

11 years agoMake tor-gencert create 2048 bit signing keys.
Linus Nordberg [Thu, 12 Dec 2013 16:39:22 +0000 (17:39 +0100)] 
Make tor-gencert create 2048 bit signing keys.

12 years agoFix get_configured_bridge_by_addr_port_digest(.,.,NULL)
rl1987 [Sun, 8 Dec 2013 18:42:33 +0000 (20:42 +0200)] 
Fix get_configured_bridge_by_addr_port_digest(.,.,NULL)

The old behavior was that NULL matched only bridges without known
identities; the correct behavior is that NULL should match all
bridges (assuming that their addr:port matches).

12 years agoMerge remote-tracking branch 'public/bug10131_024'
Nick Mathewson [Mon, 9 Dec 2013 16:06:20 +0000 (11:06 -0500)] 
Merge remote-tracking branch 'public/bug10131_024'

12 years agoRemove a check in channeltls.c that could never fail.
Nick Mathewson [Mon, 9 Dec 2013 16:02:34 +0000 (11:02 -0500)] 
Remove a check in channeltls.c that could never fail.

We were checking whether a 8-bit length field had overflowed a
503-byte buffer. Unless somebody has found a way to store "504" in a
single byte, it seems unlikely.

Fix for 10313 and 9980. Based on a pach by Jared L Wong. First found
by David Fifield with STACK.

12 years agoSet CREATE_NO_WINDOW in tor_spawn_background.
David Fifield [Thu, 5 Dec 2013 04:56:28 +0000 (04:56 +0000)] 
Set CREATE_NO_WINDOW in tor_spawn_background.

This flag prevents the creation of a console window popup on Windows. We
need it for pluggable transport executables--otherwise you get blank
console windows when you launch the 3.x browser bundle with transports
enabled.

http://msdn.microsoft.com/en-us/library/ms684863.aspx#CREATE_NO_WINDOW

The browser bundles that used Vidalia used to set this flag when
launching tor itself; it was apparently inherited by the pluggable
transports launched by tor. In the 3.x bundles, tor is launched by some
JavaScript code, which doesn't have the ability to set CREATE_NO_WINDOW.
tor itself is now being compiled with the -mwindows option, so that it
is a GUI application, not a console application, and doesn't show a
console window in any case. This workaround doesn't work for pluggable
transports, because they need to be able to write control messages to
stdout.

https://trac.torproject.org/projects/tor/ticket/9444#comment:30

12 years agoMerge branch 'bug4677'
Nick Mathewson [Mon, 25 Nov 2013 15:56:52 +0000 (10:56 -0500)] 
Merge branch 'bug4677'

12 years agoRestore prop198 behavior from 4677 patch
Nick Mathewson [Mon, 25 Nov 2013 15:53:37 +0000 (10:53 -0500)] 
Restore prop198 behavior from 4677 patch

The previous commit from piet would have backed out some of proposal
198 and made servers built without the V2 handshake not use the
unrestricted cipher list from prop198.

Bug not in any released Tor.

12 years agoRestore ability to build with V2_HANDSHAKE_SERVER
Nick Mathewson [Mon, 25 Nov 2013 15:45:20 +0000 (10:45 -0500)] 
Restore ability to build with V2_HANDSHAKE_SERVER

Fixes bug 4677; bugfix on 0.2.3.2-alpha. Fix by "piet".

12 years agoAvoid warning about impossible check for flags & 0
Nick Mathewson [Fri, 22 Nov 2013 17:42:05 +0000 (12:42 -0500)] 
Avoid warning about impossible check for flags & 0

Fixes CID 743381

12 years agoOnly update view of micrdescriptor pos if pos is fetchable.
Nick Mathewson [Fri, 22 Nov 2013 17:38:58 +0000 (12:38 -0500)] 
Only update view of micrdescriptor pos if pos is fetchable.

It's conceivable (but probably impossible given our code) that lseek
could return -1 on an error; when that happens, we don't want off to
become -1.

Fixes CID 1035124.

12 years agoAdd checks to prevent memcmp(.,.,negative) in tests (CID 1064417)
Nick Mathewson [Fri, 22 Nov 2013 17:33:25 +0000 (12:33 -0500)] 
Add checks to prevent memcmp(.,.,negative) in tests (CID 1064417)

12 years agoFix another unit test memory leak. CID1087949,CID1087950.
Nick Mathewson [Fri, 22 Nov 2013 17:27:41 +0000 (12:27 -0500)] 
Fix another unit test memory leak. CID1087949,CID1087950.

12 years agoFix a bunch of coverity-spotted unit test resource leaks
Nick Mathewson [Fri, 22 Nov 2013 17:21:14 +0000 (12:21 -0500)] 
Fix a bunch of coverity-spotted unit test resource leaks

CIDs: 1130994113099311309921130991

12 years agoRemove needless fd var from test. CID 1130989.
Nick Mathewson [Fri, 22 Nov 2013 17:16:17 +0000 (12:16 -0500)] 
Remove needless fd var from test. CID 1130989.

12 years agoHandle unlikely negative time in tor_log_err_sigsafe
Nick Mathewson [Fri, 22 Nov 2013 17:14:11 +0000 (12:14 -0500)] 
Handle unlikely negative time in tor_log_err_sigsafe

Coverity wants this; CID 1130990.

12 years agoWhitespace cleanup
Nick Mathewson [Wed, 20 Nov 2013 16:05:00 +0000 (11:05 -0500)] 
Whitespace cleanup

12 years agoMerge remote-tracking branch 'sysrqb/bug9859_5'
Nick Mathewson [Wed, 20 Nov 2013 16:03:37 +0000 (11:03 -0500)] 
Merge remote-tracking branch 'sysrqb/bug9859_5'

12 years agoFix crypto/digests test
Nick Mathewson [Mon, 18 Nov 2013 18:33:29 +0000 (13:33 -0500)] 
Fix crypto/digests test

12 years agoMerge branch 'finish_prop157'
Nick Mathewson [Mon, 18 Nov 2013 18:27:06 +0000 (13:27 -0500)] 
Merge branch 'finish_prop157'

12 years agoTweak #10162 documentation a bit
Nick Mathewson [Mon, 18 Nov 2013 18:26:58 +0000 (13:26 -0500)] 
Tweak #10162 documentation a bit

12 years agoAdd a _GNU_SOURCE definition to backtrace.c to fix compilation
Nick Mathewson [Mon, 18 Nov 2013 18:05:23 +0000 (13:05 -0500)] 
Add a _GNU_SOURCE definition to backtrace.c to fix compilation

12 years agoWhoops; changes files belong in changes.
Nick Mathewson [Mon, 18 Nov 2013 17:49:12 +0000 (12:49 -0500)] 
Whoops; changes files belong in changes.

12 years agoWhoops -- add missing defined().
Nick Mathewson [Mon, 18 Nov 2013 16:36:23 +0000 (11:36 -0500)] 
Whoops -- add missing defined().

12 years agoMake header includes match declarations in pc_from_ucontext.m4
Nick Mathewson [Mon, 18 Nov 2013 16:34:15 +0000 (11:34 -0500)] 
Make header includes match declarations in pc_from_ucontext.m4

With any luck, this will clean up errors where we detect that
REG_{EIP,RIP} is present in autoconf, but when we go to include it,
it isn't there.

12 years agoFix compilation under openssl 0.9.8
Nick Mathewson [Mon, 18 Nov 2013 16:23:54 +0000 (11:23 -0500)] 
Fix compilation under openssl 0.9.8

It's not nice to talk about NID_aes_{128,256}_{ctr,gcm} when they
don't exist.

Fix on 84458b79a78ea7e26820bf0; bug not in any released Tor.

12 years agoRemove 'struct timeval now' that was shadowing 'struct timeval now'.
Nick Mathewson [Mon, 18 Nov 2013 16:20:35 +0000 (11:20 -0500)] 
Remove 'struct timeval now' that was shadowing 'struct timeval now'.

This was a mistake in the merge commit 7a2b30fe16eacc040b3dd11. It
would have made the CellStatistics code give completely bogus
results. Bug not in any released Tor.

12 years agoFix whitespace
Nick Mathewson [Mon, 18 Nov 2013 16:13:40 +0000 (11:13 -0500)] 
Fix whitespace

12 years agoLog more OpenSSL engine statuses at startup.
Nick Mathewson [Mon, 18 Nov 2013 16:12:24 +0000 (11:12 -0500)] 
Log more OpenSSL engine statuses at startup.

Fixes ticket 10043; patch from Joshua Datko.

12 years agoMerge branch 'backtrace_squashed_merged'
Nick Mathewson [Mon, 18 Nov 2013 16:01:31 +0000 (11:01 -0500)] 
Merge branch 'backtrace_squashed_merged'

12 years agoMerge branch 'backtrace_squashed'
Nick Mathewson [Mon, 18 Nov 2013 16:00:16 +0000 (11:00 -0500)] 
Merge branch 'backtrace_squashed'

Conflicts:
src/common/sandbox.c
src/common/sandbox.h
src/common/util.c
src/or/main.c
src/test/include.am
src/test/test.c

12 years agoImprove backtrace changes file
Nick Mathewson [Mon, 18 Nov 2013 15:48:08 +0000 (10:48 -0500)] 
Improve backtrace changes file

12 years agoRefactor format_*_number_sigsafe to have a common implementation
Nick Mathewson [Fri, 18 Oct 2013 17:37:23 +0000 (10:37 -0700)] 
Refactor format_*_number_sigsafe to have a common implementation

12 years agoReseolve DOCDOC and XXXXs in backtrace.c
Nick Mathewson [Thu, 15 Aug 2013 16:45:46 +0000 (12:45 -0400)] 
Reseolve DOCDOC and XXXXs in backtrace.c

12 years agoTests for backtrace.c
Nick Mathewson [Tue, 30 Jul 2013 01:56:31 +0000 (21:56 -0400)] 
Tests for backtrace.c

These need to be a separate executable, since the point of backtrace.c
is that it can crash and write stuff.

12 years agoUnit tests for new functions in log.c
Nick Mathewson [Mon, 29 Jul 2013 17:30:49 +0000 (13:30 -0400)] 
Unit tests for new functions in log.c

12 years agoMake backtrace handler handle signals correctly.
Nick Mathewson [Fri, 26 Jul 2013 11:22:56 +0000 (13:22 +0200)] 
Make backtrace handler handle signals correctly.

This meant moving a fair bit of code around, and writing a signal
cleanup function.  Still pretty nice from what I can tell, though.

12 years agoUse pc_from_ucontext.m4 from Google Performance Tools
Nick Mathewson [Fri, 26 Jul 2013 10:46:17 +0000 (12:46 +0200)] 
Use pc_from_ucontext.m4 from Google Performance Tools

This M4 module lets us learn the right way (out of at least 18
possibilities) to extract the current PC for stack-trace-fixup-in-signal
purposes.  The Google Performance Tools license is 3-clause BSD.

12 years agoImprove new assertion message logging
Nick Mathewson [Thu, 25 Jul 2013 10:12:35 +0000 (12:12 +0200)] 
Improve new assertion message logging

Don't report that a failure happened in the assertion_failed function just
because we logged it from there.

12 years agoAdd a sighandler-safe logging mechanism
Nick Mathewson [Sat, 20 Jul 2013 02:47:49 +0000 (22:47 -0400)] 
Add a sighandler-safe logging mechanism

We had accidentially grown two fake ones: one for backtrace.c, and one
for sandbox.c.  Let's do this properly instead.

Now, when we configure logs, we keep track of fds that should get told
about bad stuff happening from signal handlers.  There's another entry
point for these that avoids using non-signal-handler-safe functions.

12 years agoOn Linux (and some other systems) we need -rdynamic for backtraces
Nick Mathewson [Fri, 19 Jul 2013 18:18:43 +0000 (14:18 -0400)] 
On Linux (and some other systems) we need -rdynamic for backtraces

12 years agoBasic backtrace ability
Nick Mathewson [Fri, 19 Jul 2013 18:09:58 +0000 (14:09 -0400)] 
Basic backtrace ability

On platforms with the backtrace/backtrace_symbols_fd interface, Tor
can now dump stack traces on assertion failure.  By default, I log
them to DataDir/stack_dump and to stderr.

12 years agoforward-port the 0.2.4.18-rc changelog
Roger Dingledine [Sun, 17 Nov 2013 04:02:26 +0000 (23:02 -0500)] 
forward-port the 0.2.4.18-rc changelog

12 years agoMerge remote-tracking branch 'origin/maint-0.2.4'
Nick Mathewson [Fri, 15 Nov 2013 20:35:00 +0000 (15:35 -0500)] 
Merge remote-tracking branch 'origin/maint-0.2.4'

Conflicts:
src/or/relay.c

Conflict changes were easy; compilation fixes required were using
using TOR_SIMPLEQ_FIRST to get head of cell queue.

12 years agoMerge remote-tracking branch 'origin/maint-0.2.3' into maint-0.2.4
Nick Mathewson [Fri, 15 Nov 2013 20:29:24 +0000 (15:29 -0500)] 
Merge remote-tracking branch 'origin/maint-0.2.3' into maint-0.2.4

Conflicts:
src/or/or.h
src/or/relay.c

Conflicts were simple to resolve.  More fixes were needed for
compilation, including: reinstating the tv_to_msec function, and renaming
*_conn_cells to *_chan_cells.

12 years agoMerge branch 'bug9093_023' into maint-0.2.3
Nick Mathewson [Fri, 15 Nov 2013 20:23:51 +0000 (15:23 -0500)] 
Merge branch 'bug9093_023' into maint-0.2.3

12 years agoMake the dir-key-crosscert element required
Nick Mathewson [Thu, 14 Nov 2013 14:37:41 +0000 (09:37 -0500)] 
Make the dir-key-crosscert element required

In proposal 157, we added a cross-certification element for
directory authority certificates. We implemented it in
0.2.1.9-alpha.  All Tor directory authorities now generate it.
Here, as planned, make it required, so that we can finally close
proposal 157.

The biggest change in the code is in the unit test data, where some
old hardcoded certs that we made long ago have become no longer
valid and now need to be replaced.

12 years agoFix test_cmdline_args to work with old openssl
Nick Mathewson [Mon, 11 Nov 2013 20:22:08 +0000 (15:22 -0500)] 
Fix test_cmdline_args to work with old openssl

If openssl was old, Tor would add a warning about its version in
between saying "no torrc found, using reasonable defaults" and
"configuration was valid".

12 years agoMerge remote-tracking branch 'origin/maint-0.2.4'
Nick Mathewson [Mon, 11 Nov 2013 15:50:05 +0000 (10:50 -0500)] 
Merge remote-tracking branch 'origin/maint-0.2.4'

12 years agomissing changes file for #10124
Nick Mathewson [Mon, 11 Nov 2013 15:49:44 +0000 (10:49 -0500)] 
missing changes file for #10124

12 years agoMerge remote-tracking branch 'origin/maint-0.2.4'
Nick Mathewson [Sun, 10 Nov 2013 17:24:12 +0000 (12:24 -0500)] 
Merge remote-tracking branch 'origin/maint-0.2.4'

12 years agoFix two more DirServer mentions in log
Nick Mathewson [Sun, 10 Nov 2013 17:23:56 +0000 (12:23 -0500)] 
Fix two more DirServer mentions in log

12 years agoFix a wide line
Nick Mathewson [Sun, 10 Nov 2013 17:22:34 +0000 (12:22 -0500)] 
Fix a wide line

12 years agoChange documentation DirServer->DirAuthority
Nick Mathewson [Sun, 10 Nov 2013 17:21:23 +0000 (12:21 -0500)] 
Change documentation DirServer->DirAuthority

We renamed the option, but we didn't actually fix it in the log
messages or the docs.  This patch does that.

For #10124.  Patch by sqrt2.

12 years agoTweak file documentation in tor.1.txt
Nick Mathewson [Thu, 7 Nov 2013 19:52:29 +0000 (14:52 -0500)] 
Tweak file documentation in tor.1.txt

12 years agoUpdating manpage to describe some of the data directory files.
rl1987 [Sun, 3 Nov 2013 20:27:06 +0000 (22:27 +0200)] 
Updating manpage to describe some of the data directory files.

12 years agoDocumentation and tests for 10060
Nick Mathewson [Thu, 7 Nov 2013 19:42:58 +0000 (14:42 -0500)] 
Documentation and tests for 10060

12 years agoImplementing --allow-missing-torrc CLI option.
rl1987 [Sun, 3 Nov 2013 16:53:41 +0000 (18:53 +0200)] 
Implementing --allow-missing-torrc CLI option.

12 years agoImproved circuit queue out-of-memory handler
Nick Mathewson [Thu, 7 Nov 2013 17:15:30 +0000 (12:15 -0500)] 
Improved circuit queue out-of-memory handler

Previously, when we ran low on memory, we'd close whichever circuits
had the most queued cells. Now, we close those that have the
*oldest* queued cells, on the theory that those are most responsible
for us running low on memory, and that those are the least likely to
actually drain on their own if we wait a little longer.

Based on analysis from a forthcoming paper by Jansen, Tschorsch,
Johnson, and Scheuermann. Fixes bug 9093.

12 years agoMerge branch 'prop221_squashed_024'
Nick Mathewson [Fri, 1 Nov 2013 14:28:01 +0000 (10:28 -0400)] 
Merge branch 'prop221_squashed_024'

Conflicts:
src/or/or.h

12 years agocircuit_build_failed: distinguish "got DESTROY" case
Nick Mathewson [Thu, 31 Oct 2013 20:53:31 +0000 (16:53 -0400)] 
circuit_build_failed: distinguish "got DESTROY" case

Roger spotted this on tor-dev in his comments on proposal 221.

We etect DESTROY vs everything else, since arma likes network
timeout indicating failure but not overload indicating failure.

12 years agoImplement proposal 221: Stop sending CREATE_FAST
Nick Mathewson [Thu, 31 Oct 2013 20:44:14 +0000 (16:44 -0400)] 
Implement proposal 221: Stop sending CREATE_FAST

This makes FastFirstHopPK an AUTOBOOL; makes the default "auto"; and
makes the behavior of "auto" be "look at the consensus."

12 years agoMerge remote-tracking branch 'origin/maint-0.2.4'
Nick Mathewson [Thu, 31 Oct 2013 20:10:19 +0000 (16:10 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.4'

12 years agoMerge remote-tracking branch 'public/bug9645' into maint-0.2.4
Nick Mathewson [Thu, 31 Oct 2013 20:09:41 +0000 (16:09 -0400)] 
Merge remote-tracking branch 'public/bug9645' into maint-0.2.4

12 years agoAppease make check-spaces
Nick Mathewson [Thu, 31 Oct 2013 18:45:20 +0000 (14:45 -0400)] 
Appease make check-spaces

12 years agoMerge remote-tracking branch 'public/bug10063'
Nick Mathewson [Thu, 31 Oct 2013 18:43:20 +0000 (14:43 -0400)] 
Merge remote-tracking branch 'public/bug10063'

12 years agoAdd missing includes for circpathbias.h
Nick Mathewson [Thu, 31 Oct 2013 18:33:34 +0000 (14:33 -0400)] 
Add missing includes for circpathbias.h

12 years agoMove other #9731 check to start of its functions
Nick Mathewson [Thu, 31 Oct 2013 18:10:23 +0000 (14:10 -0400)] 
Move other #9731 check to start of its functions

At arma's suggestion.  Looks like I missed this one.

12 years agoMake circpathbias and circuitbuild compile.
Nick Mathewson [Thu, 31 Oct 2013 18:28:49 +0000 (14:28 -0400)] 
Make circpathbias and circuitbuild compile.

That was the tricky part

12 years agoMove pathbias functions into a new file.
Nick Mathewson [Thu, 31 Oct 2013 18:17:32 +0000 (14:17 -0400)] 
Move pathbias functions into a new file.

Does not compile yet.  This is the "no code changed" diff.

12 years agoMove #9731 checks to start of their functions
Nick Mathewson [Thu, 31 Oct 2013 18:10:23 +0000 (14:10 -0400)] 
Move #9731 checks to start of their functions

At arma's suggestion.

12 years agoMerge remote-tracking branch 'public/bug9731'
Nick Mathewson [Thu, 31 Oct 2013 18:09:18 +0000 (14:09 -0400)] 
Merge remote-tracking branch 'public/bug9731'

12 years agoMerge remote-tracking branch 'origin/maint-0.2.4'
Nick Mathewson [Thu, 31 Oct 2013 18:08:57 +0000 (14:08 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.4'

Do an "ours" merge to not take the 0.2.4 fix for #9731

12 years agoMerge remote-tracking branch 'public/bug9731b' into maint-0.2.4
Nick Mathewson [Thu, 31 Oct 2013 18:08:28 +0000 (14:08 -0400)] 
Merge remote-tracking branch 'public/bug9731b' into maint-0.2.4

12 years agoMerge remote-tracking branch 'origin/maint-0.2.4'
Nick Mathewson [Thu, 31 Oct 2013 18:03:01 +0000 (14:03 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.4'

12 years agoMerge remote-tracking branch 'public/bug9780_024_v2' into maint-0.2.4
Nick Mathewson [Thu, 31 Oct 2013 18:02:28 +0000 (14:02 -0400)] 
Merge remote-tracking branch 'public/bug9780_024_v2' into maint-0.2.4

12 years agoMerge remote-tracking branch 'origin/maint-0.2.4'
Nick Mathewson [Thu, 31 Oct 2013 17:26:33 +0000 (13:26 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.4'

12 years agoMerge remote-tracking branch 'public/bug6055_v2_024' into maint-0.2.4
Nick Mathewson [Thu, 31 Oct 2013 17:14:39 +0000 (13:14 -0400)] 
Merge remote-tracking branch 'public/bug6055_v2_024' into maint-0.2.4

12 years agoOnly use -Wlogical-op with GCC 4.6 and later
Nick Mathewson [Thu, 31 Oct 2013 15:57:50 +0000 (11:57 -0400)] 
Only use -Wlogical-op with GCC 4.6 and later

It generates an apparently spurious warning with gcc 4.4 in debian;
we haven't tested 4.5.

12 years agoMention that tor is client-by-default. From Weasel. Resolves #10057
Nick Mathewson [Thu, 31 Oct 2013 15:08:21 +0000 (11:08 -0400)] 
Mention that tor is client-by-default. From Weasel. Resolves #10057

12 years agoMinor manpage tweaks from weasel; closes #10058
Nick Mathewson [Thu, 31 Oct 2013 14:55:18 +0000 (10:55 -0400)] 
Minor manpage tweaks from weasel; closes #10058

12 years agoAdd changelog entry for merged #7359.
Karsten Loesing [Thu, 31 Oct 2013 08:21:13 +0000 (09:21 +0100)] 
Add changelog entry for merged #7359.

12 years agoAdd some clarity and checks to cell_queue_append_packed_copy
Nick Mathewson [Thu, 31 Oct 2013 03:00:52 +0000 (23:00 -0400)] 
Add some clarity and checks to cell_queue_append_packed_copy

It's not cool to have "circ may be NULL if use_stats false, but
otherwise we crash" as an undocumented API constraint. :)

12 years agoMerge remote-tracking branch 'karsten/morestats5'
Nick Mathewson [Thu, 31 Oct 2013 02:53:05 +0000 (22:53 -0400)] 
Merge remote-tracking branch 'karsten/morestats5'

12 years agoAdd the -Wlogical-op switch when building with GCC 4.3 or later.
Nick Mathewson [Thu, 31 Oct 2013 02:39:53 +0000 (22:39 -0400)] 
Add the -Wlogical-op switch when building with GCC 4.3 or later.

12 years agoFix an always-true assert in PT code.
George Kadianakis [Tue, 29 Oct 2013 22:49:37 +0000 (22:49 +0000)] 
Fix an always-true assert in PT code.

12 years agoMerge remote-tracking branch 'public/bug5018'
Nick Mathewson [Tue, 29 Oct 2013 05:29:59 +0000 (01:29 -0400)] 
Merge remote-tracking branch 'public/bug5018'

Conflicts:
src/or/entrynodes.c