]> git.ipfire.org Git - thirdparty/tor.git/log
thirdparty/tor.git
13 years agoFix whitespace
Nick Mathewson [Mon, 27 Aug 2012 14:53:40 +0000 (10:53 -0400)] 
Fix whitespace

13 years agoMerge branch 'bug6524_nm'
Nick Mathewson [Mon, 27 Aug 2012 14:33:24 +0000 (10:33 -0400)] 
Merge branch 'bug6524_nm'

13 years agoChanges file for nonrecursive make branch
Nick Mathewson [Mon, 27 Aug 2012 14:28:45 +0000 (10:28 -0400)] 
Changes file for nonrecursive make branch

13 years agoUpdate .gitignore with new autotools droppings from nonrecursive make
Nick Mathewson [Mon, 27 Aug 2012 14:11:55 +0000 (10:11 -0400)] 
Update .gitignore with new autotools droppings from nonrecursive make

13 years agoBetter fix to avoid loudness on mkdir -p
Jim Meyering [Mon, 27 Aug 2012 14:05:02 +0000 (10:05 -0400)] 
Better fix to avoid loudness on mkdir -p

(commit message by nickm)

13 years agobuild: minimal adjustments to make out-of-tree build work
Jim Meyering [Mon, 27 Aug 2012 14:00:22 +0000 (10:00 -0400)] 
build: minimal adjustments to make out-of-tree build work

13 years agoMerge remote-tracking branch 'linus/bug6364'
Nick Mathewson [Mon, 27 Aug 2012 13:53:37 +0000 (09:53 -0400)] 
Merge remote-tracking branch 'linus/bug6364'

13 years agoMerge remote-tracking branch 'linus/bug6362'
Nick Mathewson [Mon, 27 Aug 2012 13:50:22 +0000 (09:50 -0400)] 
Merge remote-tracking branch 'linus/bug6362'

13 years agoMerge remote-tracking branch 'origin/maint-0.2.3'
Nick Mathewson [Fri, 24 Aug 2012 16:51:50 +0000 (12:51 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.3'

13 years agoMerge remote-tracking branch 'public/bug6472' into maint-0.2.3
Nick Mathewson [Fri, 24 Aug 2012 16:51:02 +0000 (12:51 -0400)] 
Merge remote-tracking branch 'public/bug6472' into maint-0.2.3

13 years agoMake node_assert_ok less duplicatey
Nick Mathewson [Fri, 24 Aug 2012 16:48:23 +0000 (12:48 -0400)] 
Make node_assert_ok less duplicatey

This comes at the cost of making its failure message a little less
friendly, but since when do assertion failures count as
user-friendly?

13 years agoClarify docs on get_configured_bridge_by_*_digest
Nick Mathewson [Fri, 24 Aug 2012 16:31:09 +0000 (12:31 -0400)] 
Clarify docs on get_configured_bridge_by_*_digest

13 years agoWe now need automake 1.9 or later
Nick Mathewson [Fri, 24 Aug 2012 14:29:05 +0000 (10:29 -0400)] 
We now need automake 1.9 or later

Automake 1.7 is too broken to even investigate why it broke at this point.

13 years agoMove ipv6_preferred from routerinfo_t to node_t.
Linus Nordberg [Thu, 23 Aug 2012 10:23:00 +0000 (12:23 +0200)] 
Move ipv6_preferred from routerinfo_t to node_t.

Move extend_info_from_router() from circuitbuild.c to router.c and
make it static.

Add get_configured_bridge_by_orports_digest() and have
get_configured_bridge_by_routerinfo() and
node_is_a_configured_bridge() use it. We now consider all OR ports of
a bridge when looking for it.

Move node_get_*_orport to nodelist.c.

Fix a cut'n'paste error in header of nodelist.h.

Add node_assert_ok().

Add router_get_all_orports(). It's duplicating code from
node_get_all_orports(). Worth fixing at the cost of complicating the
API slightly?

13 years agoAdd some attributes to make a2x quieter
Nick Mathewson [Thu, 23 Aug 2012 17:19:54 +0000 (13:19 -0400)] 
Add some attributes to make a2x quieter

13 years agoMake asciidoc generation quieter
Nick Mathewson [Thu, 23 Aug 2012 17:19:40 +0000 (13:19 -0400)] 
Make asciidoc generation quieter

13 years agoMake the _sha1.i file generation quieter
Nick Mathewson [Thu, 23 Aug 2012 17:14:41 +0000 (13:14 -0400)] 
Make the _sha1.i file generation quieter

13 years agoAdd missing dependency so the html.in files get built
Nick Mathewson [Thu, 23 Aug 2012 16:52:33 +0000 (12:52 -0400)] 
Add missing dependency so the html.in files get built

13 years agoReplace man_MANS with nodist_man1_MANS
Jim Meyering [Thu, 23 Aug 2012 16:36:31 +0000 (12:36 -0400)] 
Replace man_MANS with nodist_man1_MANS

(commit message by nickm)

13 years agos/AM_CONFIG_HEADER/AC_CONFIG_HEADERS/
Jim Meyering [Thu, 23 Aug 2012 16:35:21 +0000 (12:35 -0400)] 
s/AM_CONFIG_HEADER/AC_CONFIG_HEADERS/

Apparently the former will stop working with newer auto*

(commit message by nickm)

13 years agoRemove #6480 fix from 0.2.2.38 changelog - it was not included
Peter Palfrader [Wed, 22 Aug 2012 15:51:01 +0000 (17:51 +0200)] 
Remove #6480 fix from 0.2.2.38 changelog - it was not included

(Forward-port from release-0.2.2 to master)

13 years agoFetch IPv6 address from NETINFO "other OR's address" field.
Linus Nordberg [Tue, 21 Aug 2012 16:43:36 +0000 (18:43 +0200)] 
Fetch IPv6 address from NETINFO "other OR's address" field.

The my_apparent_addr is still unused, apart from now being logged in
the "Got good NETINFO cell" info message.

13 years agoMerge branch 'bug6638'
Nick Mathewson [Tue, 21 Aug 2012 15:20:58 +0000 (11:20 -0400)] 
Merge branch 'bug6638'

13 years agoMerge remote-tracking branch 'origin/maint-0.2.3'
Nick Mathewson [Tue, 21 Aug 2012 14:36:51 +0000 (10:36 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.3'

13 years agoMerge remote-tracking branch 'public/bug6404' into maint-0.2.3
Nick Mathewson [Tue, 21 Aug 2012 14:35:40 +0000 (10:35 -0400)] 
Merge remote-tracking branch 'public/bug6404' into maint-0.2.3

13 years agoMerge remote-tracking branch 'linus/bug6621'
Nick Mathewson [Tue, 21 Aug 2012 14:21:53 +0000 (10:21 -0400)] 
Merge remote-tracking branch 'linus/bug6621'

13 years agoAvoid spurious warnings in rend_service_introduce
Nick Mathewson [Tue, 21 Aug 2012 14:15:52 +0000 (10:15 -0400)] 
Avoid spurious warnings in rend_service_introduce

There was some code in the "err:" block that would always log a
warning, reporting an "unknown error" if we hadn't set err_msg.  But
there were also plenty of "goto err" blocks that did their own
logging, and never set err_msg at all.  Now we should only log when
we have an error message to log.

This fixes bug 6638, from no released Tor version.

13 years agoSend IPv6 address in NETINFO cells.
Linus Nordberg [Tue, 14 Aug 2012 12:03:58 +0000 (14:03 +0200)] 
Send IPv6 address in NETINFO cells.

Closes #6364.

13 years agoMake all relays, not only bridges, capable of advertising an IPv6 OR port.
Linus Nordberg [Tue, 14 Aug 2012 12:03:58 +0000 (14:03 +0200)] 
Make all relays, not only bridges, capable of advertising an IPv6 OR port.

Closes #6362.

13 years agoMerge remote-tracking branch 'origin/maint-0.2.3'
Nick Mathewson [Fri, 17 Aug 2012 21:10:16 +0000 (17:10 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.3'

13 years agowhitespace fix
Nick Mathewson [Fri, 17 Aug 2012 21:10:03 +0000 (17:10 -0400)] 
whitespace fix

13 years agoWhen iterating over connections pending DNS, skip marked ones
Nick Mathewson [Fri, 17 Aug 2012 20:46:11 +0000 (16:46 -0400)] 
When iterating over connections pending DNS, skip marked ones

Failure to do this would lead to double-free cases and similar,
especially when the exit's DNS was broken. See bug 6472 for full
details; this is a fix for 6472.

Anonymous patch from "cypherpunks" on trac.

13 years agoMerge remote-tracking branch 'public/bug5124'
Nick Mathewson [Fri, 17 Aug 2012 20:05:09 +0000 (16:05 -0400)] 
Merge remote-tracking branch 'public/bug5124'

13 years agoMerge branch 'bug6468_squashed'
Nick Mathewson [Fri, 17 Aug 2012 20:01:43 +0000 (16:01 -0400)] 
Merge branch 'bug6468_squashed'

13 years agoRemove needless flush-on-write code.
Nick Mathewson [Wed, 25 Jul 2012 22:34:14 +0000 (18:34 -0400)] 
Remove needless flush-on-write code.

Long ago, before we had cell queues, it was necessary to maybe call
connection_handle_write() from connectino_write_to_buf_impl() on OR
connections, so that we wouldn't get into a loop of reading infinite
amounts of data and queueing it all on an outbuf before bothering to
write any data.

If that doesn't sounds like what our code does now, you're right:
right now, we won't stick more than OR_CONN_HIGHWATER bytes of cells
on an outbuf, and we won't suck more than CELL_QUEUE_HIGHWATER_SIZE
cells off any edge connection. So, there's no more call for that
code.

Removing this code will simplify our data flow, and that should be
something we can all get behind.

13 years agoMerge remote-tracking branch 'origin/maint-0.2.3'
Nick Mathewson [Fri, 17 Aug 2012 19:59:40 +0000 (15:59 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.3'

13 years agoConsider IPv6 OR ports when deciding whether a routerinfo change is cosmetic.
Linus Nordberg [Thu, 19 Jul 2012 18:41:23 +0000 (20:41 +0200)] 
Consider IPv6 OR ports when deciding whether a routerinfo change is cosmetic.

Closes #6423.

13 years agoMerge remote-tracking branch 'origin/maint-0.2.3'
Nick Mathewson [Fri, 17 Aug 2012 18:04:28 +0000 (14:04 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.3'

13 years agoMerge branch 'bug6379' into maint-0.2.3
Nick Mathewson [Fri, 17 Aug 2012 17:57:07 +0000 (13:57 -0400)] 
Merge branch 'bug6379' into maint-0.2.3

13 years agoFix more warnings from openbsd_malloc
Nick Mathewson [Fri, 17 Aug 2012 17:49:52 +0000 (13:49 -0400)] 
Fix more warnings from openbsd_malloc

Apparently, (void)writev is not enough to suppress the "you are
ignoring the return value!" warnings on Linux.  Instead, remove the
whole warning/error logic when compiling openbsd_malloc for Tor: we
can't use it.

13 years agoMerge remote-tracking branch 'origin/maint-0.2.3'
Nick Mathewson [Fri, 17 Aug 2012 16:39:31 +0000 (12:39 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.3'

13 years agoMerge remote-tracking branch 'public/bug6244_part_c' into maint-0.2.3
Nick Mathewson [Fri, 17 Aug 2012 16:37:49 +0000 (12:37 -0400)] 
Merge remote-tracking branch 'public/bug6244_part_c' into maint-0.2.3

13 years agoMerge remote-tracking branch 'origin/maint-0.2.3'
Nick Mathewson [Fri, 17 Aug 2012 16:34:52 +0000 (12:34 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.3'

13 years agoMerge remote-tracking branch 'public/bug6507' into maint-0.2.3
Nick Mathewson [Fri, 17 Aug 2012 16:33:17 +0000 (12:33 -0400)] 
Merge remote-tracking branch 'public/bug6507' into maint-0.2.3

13 years agoMerge remote-tracking branch 'origin/maint-0.2.3'
Nick Mathewson [Fri, 17 Aug 2012 16:10:44 +0000 (12:10 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.3'

13 years agoWhitespace and build fixes on 6475 patch
Nick Mathewson [Fri, 17 Aug 2012 16:10:31 +0000 (12:10 -0400)] 
Whitespace and build fixes on 6475 patch

13 years agoMerge remote-tracking branch 'mikeperry/bug6475' into maint-0.2.3
Nick Mathewson [Fri, 17 Aug 2012 16:08:42 +0000 (12:08 -0400)] 
Merge remote-tracking branch 'mikeperry/bug6475' into maint-0.2.3

13 years agoMerge remote-tracking branch 'origin/maint-0.2.3'
Nick Mathewson [Fri, 17 Aug 2012 15:54:05 +0000 (11:54 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.3'

13 years agoMerge remote-tracking branch 'public/bug6514' into maint-0.2.3
Nick Mathewson [Fri, 17 Aug 2012 15:53:06 +0000 (11:53 -0400)] 
Merge remote-tracking branch 'public/bug6514' into maint-0.2.3

13 years agoMake authorities not set Running unless all advertised OR ports are reachable.
Linus Nordberg [Fri, 17 Aug 2012 10:53:25 +0000 (12:53 +0200)] 
Make authorities not set Running unless all advertised OR ports are reachable.

Resolves #6621.

13 years agoAddress Nick's comments from code review.
Mike Perry [Thu, 16 Aug 2012 23:29:19 +0000 (16:29 -0700)] 
Address Nick's comments from code review.

Also promote log messages to notice and rate-limit them.

13 years agoBug 6475: Explicitly track our path bias state.
Mike Perry [Thu, 16 Aug 2012 02:59:55 +0000 (19:59 -0700)] 
Bug 6475: Explicitly track our path bias state.

This is done to avoid spurious warns. Additional log lines are also
added to try to track down the codepaths where we are somehow overcounting
success counts.

13 years agoFix warnings and 64-bit problems in openbsd-malloc code
Nick Mathewson [Wed, 15 Aug 2012 23:26:53 +0000 (19:26 -0400)] 
Fix warnings and 64-bit problems in openbsd-malloc code

The warning fixes are:
  - Only define issetugid if it's missing.
  - Explicitly ignore the return value of writev.
  - Explicitly cast the retval of readlink() to int.

The 64-bit problems are related to just storing a size_t in an int. Not cool!  Use a size_t instead.

Fix for bug 6379. Bugfix on 0.2.0.20-rc, which introduced openbsd-malloc.

13 years agoFix wildcarded address mappings from the control port
Nick Mathewson [Wed, 15 Aug 2012 21:59:30 +0000 (17:59 -0400)] 
Fix wildcarded address mappings from the control port

Apparently, we weren't actually detecting wildcardedness when parsing
them: whoops!

bug 6244.  Bugfix on 0.2.3.9-alpha

13 years agoRaise the part of torrc mapaddress handling that knows wildcards
Nick Mathewson [Wed, 15 Aug 2012 21:49:18 +0000 (17:49 -0400)] 
Raise the part of torrc mapaddress handling that knows wildcards

This patch extracts the inner part of config_register_addressmaps --
the part that knows about detecting wildcard addresses addresses --
and makes it into a new function.  The new function is deliberately
not moved or reindented, so that the diff is smaller.

I need this to fix bug 6244.

13 years agoCheck for stream_id, not conn, on extend cells.
Nick Mathewson [Wed, 15 Aug 2012 17:16:41 +0000 (13:16 -0400)] 
Check for stream_id, not conn, on extend cells.

Extend cells aren't allowed to have a stream_id, but we were only
blocking them when they had a stream_id that corresponded to a
connection.  As far as I can tell, this change is harmless: it will
make some kinds of broken clients not work any more, but afaik nobody
actually make a client that was broken in that way.

Found while hunting for other places where we made the same mistake
as in 6271.

Bugfix on d7f50337c14c back from May 2003, which introduced
telescoping circuit construction into 0.0.2pre8.

13 years agoUpdate description of what we did to upper limit on md size
Nick Mathewson [Tue, 14 Aug 2012 07:10:14 +0000 (03:10 -0400)] 
Update description of what we did to upper limit on md size

Spotted by asn

13 years agoFix memory leak in dirvote_create_microdescriptor
Nick Mathewson [Tue, 14 Aug 2012 07:06:47 +0000 (03:06 -0400)] 
Fix memory leak in dirvote_create_microdescriptor

Found by George, who gets a cookie.

13 years agoClarify the last changes message
Nick Mathewson [Mon, 13 Aug 2012 18:50:29 +0000 (14:50 -0400)] 
Clarify the last changes message

13 years agooops: changes file for f45cde05f934ee95e0002ff7d72f7006ce156a4c
Nick Mathewson [Mon, 13 Aug 2012 18:37:53 +0000 (14:37 -0400)] 
oops: changes file for f45cde05f934ee95e0002ff7d72f7006ce156a4c

13 years agoRemove tor_malloc_roundup().
Nick Mathewson [Mon, 13 Aug 2012 17:27:32 +0000 (13:27 -0400)] 
Remove tor_malloc_roundup().

This function never actually did us any good, and it added a little
complexity.  See the changes file for more info.

13 years agoforward-port the 0.2.2.38 changelog
Roger Dingledine [Mon, 13 Aug 2012 04:35:37 +0000 (00:35 -0400)] 
forward-port the 0.2.2.38 changelog

13 years agoReject attempts to say FooPort and FooPort 0 in the same cfg domain
Nick Mathewson [Thu, 9 Aug 2012 20:13:03 +0000 (16:13 -0400)] 
Reject attempts to say FooPort and FooPort 0 in the same cfg domain

13 years agoFix spaces from last patch
Nick Mathewson [Thu, 9 Aug 2012 20:02:57 +0000 (16:02 -0400)] 
Fix spaces from last patch

13 years agoDon't infer we have a FooPort from the presence of a FooPort line
Nick Mathewson [Thu, 9 Aug 2012 19:48:43 +0000 (15:48 -0400)] 
Don't infer we have a FooPort from the presence of a FooPort line

Thanks to the changes we started making with SocksPort and friends
in 0.2.3.3-alpha, any of our code that did "if (options->Sockport)"
became wrong, since "SocksPort 0" would make that test true whereas
using the default SocksPort value would make it false.  (We didn't
actually do "if (options->SockPort)" but we did have tests for
TransPort.  When we moved DirPort, ORPort, and ControlPort over to
the same system in 0.2.3.9-alpha, the problem got worse, since our
code is littered with checks for DirPort and ORPort as booleans.

This code renames the current linelist-based FooPort options to
FooPort_lines, and adds new FooPort_set options which get set at
parse-and-validate time on the or_options_t.  FooPort_set is true
iff we will actually try to open a listener of the given type. (I
renamed the FooPort options rather than leave them alone so that
every previous user of a FooPort would need to get inspected, and so
that any new code that forgetfully uses FooPort will need fail to
compile.)

Fix for bug 6507.

13 years agofix up calling of config.status to generate docs
Stewart Smith [Fri, 3 Aug 2012 01:26:21 +0000 (11:26 +1000)] 
fix up calling of config.status to generate docs

13 years agoFix up make distcheck and greatly simplify docs dependencies (although it's still...
Stewart Smith [Fri, 3 Aug 2012 01:20:52 +0000 (11:20 +1000)] 
Fix up make distcheck and greatly simplify docs dependencies (although it's still a bit odd)

13 years agofix circular dependency for generating code digests
Stewart Smith [Thu, 2 Aug 2012 23:07:27 +0000 (09:07 +1000)] 
fix circular dependency for generating code digests

13 years agofix TESTS to include full path to src/test/test
Stewart Smith [Thu, 2 Aug 2012 23:05:18 +0000 (09:05 +1000)] 
fix TESTS to include full path to src/test/test

13 years agofix dependencies for some generated files
Stewart Smith [Thu, 2 Aug 2012 23:04:48 +0000 (09:04 +1000)] 
fix dependencies for some generated files

13 years agoremove contrib/Makefile and contrib/suse/Makefile from configure.in
Stewart Smith [Thu, 2 Aug 2012 22:59:32 +0000 (08:59 +1000)] 
remove contrib/Makefile and contrib/suse/Makefile from configure.in

(otherwise would throw an error of "required file X" not found as part
of autogen.sh)

13 years agoadd subdir-objects to AUTOMAKE_OPTIONS, this builds object files in subdirs with...
Stewart Smith [Thu, 2 Aug 2012 14:19:58 +0000 (00:19 +1000)] 
add subdir-objects to AUTOMAKE_OPTIONS, this builds object files in subdirs with non-recursive make

13 years agoMove to non-recursive make
Stewart Smith [Thu, 2 Aug 2012 13:31:43 +0000 (23:31 +1000)] 
Move to non-recursive make

This gives us a few benefits:
1) make -j clean all
   this will start working, as it should. It currently doesn't.
2) increased parallel build
   recursive make will max out at number of files in a directory,
   non-recursive make doesn't have such a limitation
3) Removal of duplicate information in make files,
   less error prone

I've also slightly updated how we call AM_INIT_AUTOMAKE, as the way
that was used was not only deprecated but will be *removed* in the next
major automake release (1.13).... so probably best that we can continue
to bulid tor without requiring old automake.
(see http://www.gnu.org/software/automake/manual/html_node/Public-Macros.html )

For more reasons  why, see resources such as:
http://miller.emu.id.au/pmiller/books/rmch/

13 years agoTemporarily make spurious sendmes warn louder at arma's suggestion.
Nick Mathewson [Thu, 9 Aug 2012 14:55:33 +0000 (10:55 -0400)] 
Temporarily make spurious sendmes warn louder at arma's suggestion.

13 years agoMerge remote-tracking branch 'origin/maint-0.2.3'
Nick Mathewson [Thu, 9 Aug 2012 14:52:45 +0000 (10:52 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.3'

13 years agoMerge remote-tracking branch 'public/bug6252_again' into maint-0.2.3
Nick Mathewson [Thu, 9 Aug 2012 14:50:11 +0000 (10:50 -0400)] 
Merge remote-tracking branch 'public/bug6252_again' into maint-0.2.3

13 years agoforward-port the 0.2.3.20-rc changelog
Roger Dingledine [Tue, 7 Aug 2012 21:27:03 +0000 (17:27 -0400)] 
forward-port the 0.2.3.20-rc changelog

13 years agoSpeak not the name of INT_MIN; it can upset older compilers
Nick Mathewson [Fri, 3 Aug 2012 17:54:12 +0000 (13:54 -0400)] 
Speak not the name of INT_MIN; it can upset older compilers

And more to the point, some GCCs will warn that you can't say it
before C90.

Bug not in any released version of Tor.

13 years agoMerge remote-tracking branch 'origin/maint-0.2.3'
Nick Mathewson [Fri, 3 Aug 2012 16:04:36 +0000 (12:04 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.3'

13 years agoMerge remote-tracking branch 'origin/maint-0.2.2' into maint-0.2.3
Nick Mathewson [Fri, 3 Aug 2012 16:04:11 +0000 (12:04 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.2' into maint-0.2.3

Conflicts:
src/or/routerlist.c

13 years agoTry to clarify impact of bug 6537
Nick Mathewson [Fri, 3 Aug 2012 15:54:11 +0000 (11:54 -0400)] 
Try to clarify impact of bug 6537

I don't personally agree that this is likely to be easy to exploit,
and some initial experimention I've done suggests that cache-miss
times are just plain too fast to get useful info out of when they're
mixed up with the rest of Tor's timing noise.  Nevertheless, I'm
leaving Robert's initial changelog entry in the git history so that he
can be the voice of reason if I'm wrong. :)

13 years agoMitigate a side-channel leak of which relays Tor chooses for a circuit
Robert Ransom [Thu, 14 Jun 2012 17:15:54 +0000 (17:15 +0000)] 
Mitigate a side-channel leak of which relays Tor chooses for a circuit

Tor's and OpenSSL's current design guarantee that there are other leaks,
but this one is likely to be more easily exploitable, and is easy to fix.

13 years agoMerge remote-tracking branch 'origin/maint-0.2.3'
Nick Mathewson [Fri, 3 Aug 2012 15:46:03 +0000 (11:46 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.3'

13 years agoRemove bogus comment claiming that an assertion is triggerable by consensus
Robert Ransom [Thu, 14 Jun 2012 15:41:11 +0000 (15:41 +0000)] 
Remove bogus comment claiming that an assertion is triggerable by consensus

13 years agoFix a bunch of "implicit 64->32" warnings from introduce refactoring
Nick Mathewson [Fri, 3 Aug 2012 15:31:04 +0000 (11:31 -0400)] 
Fix a bunch of "implicit 64->32" warnings from introduce refactoring

13 years agoMerge remote-tracking branch 'origin/maint-0.2.3'
Nick Mathewson [Fri, 3 Aug 2012 15:23:06 +0000 (11:23 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.3'

13 years agoMerge remote-tracking branch 'origin/maint-0.2.2' into maint-0.2.3
Nick Mathewson [Fri, 3 Aug 2012 15:18:40 +0000 (11:18 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.2' into maint-0.2.3

13 years agoClarify security impact of bug 6530
Nick Mathewson [Fri, 3 Aug 2012 15:16:13 +0000 (11:16 -0400)] 
Clarify security impact of bug 6530

13 years agoAvoid possible segfault when handling networkstatus vote with bad flavor
Nick Mathewson [Fri, 3 Aug 2012 14:53:00 +0000 (10:53 -0400)] 
Avoid possible segfault when handling networkstatus vote with bad flavor

Fix for 6530; fix on 0.2.2.6-alpha.

13 years agoUpdated docs for new connections.
Matthew Finkel [Thu, 2 Aug 2012 20:15:23 +0000 (16:15 -0400)] 
Updated docs for new connections.

13 years agoMerge remote-tracking branch 'sysrqb/bug6518'
Nick Mathewson [Thu, 2 Aug 2012 19:45:10 +0000 (15:45 -0400)] 
Merge remote-tracking branch 'sysrqb/bug6518'

13 years agoConstify struct sockaddr *sa parameter for check
Matthew Finkel [Thu, 2 Aug 2012 19:29:38 +0000 (15:29 -0400)] 
Constify struct sockaddr *sa parameter for check

The values are only being checked, not modified.

13 years agoRemoved redundant check_sockaddr_family_match call
Matthew Finkel [Thu, 2 Aug 2012 19:13:34 +0000 (15:13 -0400)] 
Removed redundant check_sockaddr_family_match call

13 years agoadd $(AM_V_GEN) to documentation generation Makefile targets.
Stewart Smith [Thu, 2 Aug 2012 14:05:36 +0000 (10:05 -0400)] 
add $(AM_V_GEN) to documentation generation Makefile targets.

This makes the V=1 or V=0 automake silent build options display (or hide)
the full command line used.

  GEN foo.bar

will be seen rather than the full command.

As with all automake silent rules, "make V=1" will output the full command.

13 years agoEnable the automake silent rules.
Stewart Smith [Thu, 2 Aug 2012 14:01:25 +0000 (10:01 -0400)] 
Enable the automake silent rules.

$ make V=1 # will temporarily disable them

otherwise you see:

   CC foo.c

rather than the giant long bulid line.

This makes it significantly easier to spot compiler warnings etc.

Additionally, make them conditional, so we won't error on automake <
1.11

(commits squashed by nickm.)

13 years agoDefensive programming: clear rs_out between iterations.
Nick Mathewson [Wed, 1 Aug 2012 21:25:34 +0000 (17:25 -0400)] 
Defensive programming: clear rs_out between iterations.

I can't currently find a bug here, but there are a couple of
near-misses.  Addresses ticket 6514; reported pseudonymously on
IRC.

13 years agoMerge remote-tracking branch 'origin/maint-0.2.3'
Nick Mathewson [Tue, 31 Jul 2012 21:20:36 +0000 (17:20 -0400)] 
Merge remote-tracking branch 'origin/maint-0.2.3'

13 years agoMerge branch 'bug6480_squashed' into maint-0.2.3
Nick Mathewson [Tue, 31 Jul 2012 21:19:47 +0000 (17:19 -0400)] 
Merge branch 'bug6480_squashed' into maint-0.2.3

13 years agoAvoid hard (impossible?)-to-trigger double-free in dns_resolve()
Nick Mathewson [Tue, 31 Jul 2012 16:58:19 +0000 (12:58 -0400)] 
Avoid hard (impossible?)-to-trigger double-free in dns_resolve()

Fixes 6480; fix on 0.2.0.1-alpha; based on pseudonymous patch.

13 years agoMerge branch 'bug6177_squashed'
Nick Mathewson [Tue, 31 Jul 2012 21:09:47 +0000 (17:09 -0400)] 
Merge branch 'bug6177_squashed'