Commit Graph

17283 Commits

Author SHA1 Message Date
George Kadianakis
8bb2ba13c1 Extract ARGS from SMETHOD line and attach them to transport. 2013-07-18 08:43:52 -04:00
Nick Mathewson
dd44ff663e A python script to combin gcov output with git blame 2013-07-16 15:31:07 -04:00
Nick Mathewson
f797ac465f Merge remote-tracking branch 'origin/maint-0.2.4' 2013-07-16 14:49:41 -04:00
Nick Mathewson
c36bdbd535 Re-do a cast in order to make old buggy freebsd gcc happy
Fix for #9254.  Bugfix on 0.2.4.14-alpha.

This is not actually a bug in the Tor code.
2013-07-16 14:48:12 -04:00
Nick Mathewson
4824f3ad93 Merge remote-tracking branch 'asn/bug9265' 2013-07-16 09:55:51 -04:00
Nick Mathewson
74356aaeec Remove an unused variable in test_replaycache_scrub 2013-07-16 09:53:44 -04:00
Andrea Shepard
e5d3efd892 Make contrib/coverage smarter about check file-existence edge cases 2013-07-16 06:40:08 -07:00
Andrea Shepard
459d827193 Add more replaycache.c unit tests, bringing coverage to 100% for that file 2013-07-16 06:02:22 -07:00
Andrea Shepard
9b3a166b44 Eliminate an impossible case in replaycache_scrub_if_needed_internal() 2013-07-16 06:01:50 -07:00
George Kadianakis
4ccd4b5c51 Test deeper in test_pt_parsing().
We used to test parse_{c,s}method_line() without actually testing that
the resulting transport_t was well formed.
2013-07-16 14:07:03 +03:00
Andrea Shepard
d1059a9366 Add optional target directory parameter to coverage script and add reset-gcov target to Makefile.am 2013-07-15 22:33:39 -07:00
Nick Mathewson
55f5caf096 Appease "make check-spaces" 2013-07-15 17:35:56 -04:00
Nick Mathewson
85178e2e93 Use format_hex_number_sigsafe to format syscalls in sandbox.c
This way, we don't have to use snprintf, which is not guaranteed to
be signal-safe.

(Technically speaking, strlen() and strlcpy() are not guaranteed to
be signal-safe by the POSIX standard. But I claim that they are on
every platform that supports libseccomp2, which is what matters
here.)
2013-07-15 13:07:09 -04:00
Nick Mathewson
9fda7e8cd1 Lightly refactor and test format_hex_number_sigsafe
Better tests for upper bounds, and for failing cases.

Also, change the function's interface to take a buffer length rather
than a maximum length, and then NUL-terminate: functions that don't
NUL-terminate are trouble waiting to happen.
2013-07-15 12:52:29 -04:00
Nick Mathewson
18136afbbb HEX_ERRNO_SIZE is no longer the correct upper limit for format_hex_number_sigsafe 2013-07-15 12:40:07 -04:00
Nick Mathewson
22977b7c1d Expose format_hex_number_..., and rename it to ..._sigsafe().
There are some other places in the code that will want a signal-safe
way to format numbers, so it shouldn't be static to util.c.
2013-07-15 12:26:55 -04:00
Nick Mathewson
449b2b7c58 Don't build format_helper_exit_status on win32
The only thing that used format_helper_exit_status on win32 was the
unit tests. This caused an error when we tried to leave a static
format_helper_exit_status lying around in a production object file.

The easiest solution is to admit that this way of dealing with process
exit status is Unix-only.
2013-07-15 12:17:23 -04:00
Nick Mathewson
1556b0cb12 Merge remote-tracking branch 'linus/unused_param' 2013-07-15 12:02:46 -04:00
Nick Mathewson
0f31d4f14d Merge remote-tracking branch 'linus/binshify' 2013-07-15 12:02:40 -04:00
Nick Mathewson
c0391bae75 Merge remote-tracking branch 'public/fancy_test_tricks'
Conflicts:
	src/common/include.am

Conflict was from adding testsupport.h near where sandbox.h had
already been added.
2013-07-15 12:02:18 -04:00
Linus Nordberg
3c74119be1 Traditional /bin/sh is unhappy about {a,b} globbing. 2013-07-15 16:06:39 +02:00
Linus Nordberg
d0c0f050df Avoid compiler warning 'unused param'.
Fixes #9261.
2013-07-15 15:46:36 +02:00
Peter Palfrader
2cb59be999 Fix two pre-coffee typos 2013-07-15 09:43:37 -04:00
Peter Palfrader
c9cb5a18fc And a changes file for #9258 2013-07-15 09:04:17 -04:00
Peter Palfrader
783c52b6df Reject relative control socket paths and emit a warning.
Previously we would accept relative paths, but only if they contained a
slash somewhere (not at the end).

Otherwise we would silently not work.  Closes: #9258.  Bugfix on
0.2.3.16-alpha.
2013-07-15 09:04:17 -04:00
Peter Palfrader
5cc52b242e Document get_parent_directory more 2013-07-15 09:04:17 -04:00
Roger Dingledine
6848e29307 cosmetic cleanups 2013-07-14 02:49:34 -04:00
Roger Dingledine
de7cdc0d94 put sandbox.h in the tarball, so the tarball builds 2013-07-13 20:31:18 -04:00
Nick Mathewson
aac732322a Merge remote-tracking branch 'public/gsoc-ctoader-cap-phase1-squashed' 2013-07-12 17:12:43 -04:00
Nick Mathewson
14d5e7f85e Remove a bogus semicolon spotted by Gisle Vanem 2013-07-11 15:53:35 -04:00
Cristian Toader
f9c1ba6493 Add a basic seccomp2 syscall filter on Linux
It's controlled by the new Sandbox argument.  Right now, it's rather
coarse-grained, it's Linux-only, and it may break some features.
2013-07-11 09:13:13 -04:00
Nick Mathewson
ec6c155f82 Add some basic unit tests for the circuit map data structure.
These show off the new mocking code by mocking the circuitmux code
so that we can test the circuit map code in isolation.
2013-07-10 15:26:34 -04:00
Nick Mathewson
b6e8c74667 Add rudimentary test mocking support.
This is not the most beautiful possible implementation (it requires
decorating mockable functions with ugly macros), but it actually
works, and is portable across multiple compilers and architectures.
2013-07-10 15:22:16 -04:00
Nick Mathewson
4753ad4f1d Add a script to compare directories full of gcov output
We can't just use "diff", since we don't care about changes in line
numbers, or changes in the exact number of times a line was called.
We just care about changes that make lines covered or non-coverd.  So
pre-process the files before calling diff.
2013-07-10 15:22:16 -04:00
Nick Mathewson
17e9fc09c3 Coverage support: build with --enable-coverage to have tests run with gcov
If you pass the --enable-coverage flag on the command line, we build
our testing binaries with appropriate options eo enable coverage
testing.  We also build a "tor-cov" binary that has coverage enabled,
for integration tests.

On recent OSX versions, test coverage only works with clang, not gcc.
So we warn about that.

Also add a contrib/coverage script to actually run gcov with the
appropriate options to generate useful .gcov files.  (Thanks to
automake, the .o files will not have the names that gcov expects to
find.)

Also, remove generated gcda and gcno files on clean.
2013-07-10 15:22:16 -04:00
Nick Mathewson
a3e0a87d95 Completely refactor how FILENAME_PRIVATE works
We previously used FILENAME_PRIVATE identifiers mostly for
identifiers exposed only to the unit tests... but also for
identifiers exposed to the benchmarker, and sometimes for
identifiers exposed to a similar module, and occasionally for no
really good reason at all.

Now, we use FILENAME_PRIVATE identifiers for identifiers shared by
Tor and the unit tests.  They should be defined static when we
aren't building the unit test, and globally visible otherwise. (The
STATIC macro will keep us honest here.)

For identifiers used only by the unit tests and never by Tor at all,
on the other hand, we wrap them in #ifdef TOR_UNIT_TESTS.

This is not the motivating use case for the split test/non-test
build system; it's just a test example to see how it works, and to
take a chance to clean up the code a little.
2013-07-10 15:20:10 -04:00
Nick Mathewson
f7d654b81e Start work on fancy compiler tricks to expose extra stuff to our tests
This is mainly a matter of automake trickery: we build each static
library in two versions now: one with the TOR_UNIT_TESTS macro
defined, and one without.  When TOR_UNIT_TESTS is defined, we can
enable mocking and expose more functions. When it's not defined, we
can lock the binary down more.

The alternatives would be to have alternate build modes: a "testing
configuration" for building the libraries with test support, and a
"production configuration" for building them without.  I don't favor
that approach, since I think it would mean more people runnning
binaries build for testing, or more people not running unit tests.
2013-07-10 15:20:09 -04:00
Nick Mathewson
fab99844fc Merge remote-tracking branch 'origin/maint-0.2.4' 2013-07-08 11:35:48 -04:00
Nick Mathewson
7a4145c45a Merge branch 'bug9200' into maint-0.2.4 2013-07-08 11:35:25 -04:00
Nick Mathewson
b34279d3ab Add a comment and a check for why flag indices will be <= 63 2013-07-08 11:35:06 -04:00
Nick Mathewson
dd90ab4121 Merge remote-tracking branch 'karsten/geoip-manual-update-jul2013' 2013-07-08 09:23:39 -04:00
Nick Mathewson
20634fac54 Merge remote-tracking branch 'origin/maint-0.2.4' 2013-07-08 09:23:09 -04:00
Nick Mathewson
c78c8de015 Merge remote-tracking branch 'origin/maint-0.2.3' into maint-0.2.4 2013-07-08 09:22:49 -04:00
Nick Mathewson
0b9c515870 Merge remote-tracking branch 'origin/maint-0.2.2' into maint-0.2.3 2013-07-08 09:22:00 -04:00
Karsten Loesing
2a61b0dd6b Update to the July 2013 GeoIP database. 2013-07-08 10:21:28 +02:00
Karsten Loesing
1a43dfa603 Add 2 new manual A1 substitutions. 2013-07-08 10:14:42 +02:00
Nick Mathewson
15cd79f832 FIx undefined behavior in dirvote.c
Fix a bug in the voting algorithm that could yield incorrect results
 when a non-naming authority declared too many flags. Fixes bug 9200;
 bugfix on 0.2.0.3-alpha.

Found by coverity scan.
2013-07-03 12:01:37 -04:00
Nick Mathewson
f631b73cd5 Merge remote-tracking branch 'linus/bug8530' 2013-06-29 16:02:13 -04:00
Nick Mathewson
0c3d676f9e Merge remote-tracking branch 'origin/maint-0.2.4' 2013-06-29 03:51:53 -04:00
Nick Mathewson
81a5ee77ec Merge branch 'bug9147' into maint-0.2.4 2013-06-29 03:50:37 -04:00