Commit Graph

14394 Commits

Author SHA1 Message Date
Nick Mathewson
1799c2be09 Merge remote-tracking branch 'public/bug13796' 2015-02-05 22:53:15 -05:00
Nick Mathewson
2274221557 Fix a work-counting bug introduced by the workqueue merge
David Goulet finds that when he runs a busy relay for a while with the
latest version of the git code, the number of onionskins handled
slowly dwindles to zero, with total_pending_tasks wedged at its
maximum value.

I conjecture this is because the total_pending_tasks variable isn't
decremented when we successfully cancel a job.  Fixed that.

Fixes bug 14741; bugfix not on any released version of tor.
2015-02-05 12:17:08 -05:00
Nick Mathewson
3f993dacc1 use ARRAY_LENGTH macro in domain_to_string 2015-02-05 11:01:13 -05:00
Yawning Angel
b330bdec8e Add a string representation for LD_SCHED, and a extra sanity check.
This both fixes the problem, and ensures that forgetting to update
domain_list in the future will trigger the bug codepath instead of
a NULL pointer deref.
2015-02-05 15:46:27 +00:00
Nick Mathewson
daab405168 Bump the minimum relay version to 0.2.4.18-rc
Closes #13555
2015-02-04 13:27:56 -05:00
Nick Mathewson
a8835170d7 Use getsockname, not getsockopt, on TPROXY sockets 2015-02-04 10:09:54 -05:00
Nick Mathewson
ac8f235446 fix a unit tests memory leak (my fault) 2015-02-03 15:58:48 -05:00
Nick Mathewson
5be48c5d4c Work around test_status.c weirdness
Ordinarily, get_options() can never return NULL, but with
test_status.c mocking, it can.  So test for that case.

The best fix here would be to pass the options value to a
bridge_server_mode() function.
2015-02-03 15:50:31 -05:00
Nick Mathewson
cdc49629c7 Merge branch 'bug6852'
Conflicts:
	src/or/status.c
2015-02-03 13:06:58 -05:00
Nick Mathewson
7f52dc4d03 Choose a more deliberate cutoff for clients in heartbeat 2015-02-03 13:02:22 -05:00
Nick Mathewson
d03e1da232 Merge remote-tracking branch 'public/bug9635_warnings_025'
Conflicts:
	src/test/test.c
2015-02-02 16:31:32 -05:00
Nick Mathewson
41ba4f5627 tweak based on comments from dgoulet 2015-02-02 14:42:33 -05:00
rl1987
a9caeb961f Test for 13865. 2015-02-02 14:38:36 -05:00
Nick Mathewson
5c807f30e4 Add more parenthesis to the definition of ARRAY_LENGTH 2015-02-02 14:14:35 -05:00
Nick Mathewson
03563f4723 Fix an unused-variable warning. 2015-02-02 13:35:44 -05:00
Nick Mathewson
79c7625e38 Merge branch 'feature13864_squashed' 2015-02-02 13:32:53 -05:00
rl1987
fe328d192e Allow reading torrc from stdin. 2015-02-02 13:31:56 -05:00
Nick Mathewson
69deab8b2a Merge remote-tracking branch 'public/bug13319' 2015-02-02 10:25:25 -05:00
Nick Mathewson
f4b79bc420 Merge remote-tracking branch 'sysrqb/bug14216_bad_since' 2015-02-02 10:23:52 -05:00
Nick Mathewson
55639bc67f Merge remote-tracking branch 'dgoulet/bug14202_026_v1' 2015-02-02 10:16:48 -05:00
Nick Mathewson
e78b7e2776 Merge remote-tracking branch 'public/14188_part1' 2015-02-02 10:15:26 -05:00
Matthew Finkel
4cb59ceb8e Only retry connecting to configured bridges
After connectivity problems, only try connecting to bridges which
are currently configured; don't mark bridges which we previously
used but are no longer configured.  Fixes 14216.  Reported by
and fix provided by arma.
2015-01-31 09:46:18 +00:00
Nick Mathewson
aba90b2125 Merge remote-tracking branch 'dgoulet/bug14554_026_v1' 2015-01-30 15:29:59 -05:00
David Goulet
2c41f12048 Fix: check r < 0 before checking errno
Signed-off-by: David Goulet <dgoulet@ev0ke.net>
2015-01-30 15:18:40 -05:00
David Goulet
51f793e37e Fix possible infinite loop on pipe/sock_drain()
If the returned value of read/recv is 0 (meaning EOF), we'll end up in an
infinite loop (active wait) until something is written on the pipe which is
not really what we want here especially because those functions are called
from the main thread.

Signed-off-by: David Goulet <dgoulet@ev0ke.net>
2015-01-30 15:05:18 -05:00
Nick Mathewson
097286e476 Fix some unused-argument warnings 2015-01-30 14:47:56 -05:00
Nick Mathewson
bc9ade055e Fix an uninitialized-variable warning. 2015-01-30 14:46:18 -05:00
David Goulet
44e9dafb67 Fix: test -ENOENT after config_parse_unix_port()
Check for -ENOENT instead of ENOENT after the HS port is parsed.

Signed-off-by: David Goulet <dgoulet@ev0ke.net>
2015-01-30 14:13:27 -05:00
Nick Mathewson
fac8d40886 Merge remote-tracking branch 'public/prop227_v2'
Conflicts:
	src/test/test_dir.c
2015-01-30 07:36:55 -05:00
Nick Mathewson
d1e52d9a2a Correctly handle OutboundBindAddress again.
ca5ba2956b broke this; bug not in any
released Tor.

Also fix a typo.

Fixes 14541 and 14527.  Reported by qbi.
2015-01-30 07:29:23 -05:00
Nick Mathewson
a87ea9b1c6 Merge branch 'bug14451_026_v1' 2015-01-29 15:16:15 -05:00
Nick Mathewson
f6afb04abb Use HT_BUCKET_NUM_ in more places
(patch from sysrqb)
2015-01-29 15:10:23 -05:00
Nick Mathewson
5faa017b86 Merge remote-tracking branch 'public/ticket11737' 2015-01-29 15:09:55 -05:00
Nick Mathewson
4c1a779539 Restrict unix: addresses to control and socks for now 2015-01-29 14:51:59 -05:00
Nick Mathewson
204374f7d9 Remove SocksSocket; it's now spelled differently thanks to 14451
Also, revise bug12585 changes file to mention new syntax
2015-01-29 14:46:20 -05:00
Nick Mathewson
b4a8fd8958 When there are no package lines, make consensus/packages say "".
Also, give a better error message when there is no consensus.
2015-01-29 14:14:59 -05:00
Nick Mathewson
bd630a899a Correctly reject packages lines with empty entries 2015-01-29 14:09:57 -05:00
Nick Mathewson
f935ee2dae Define 'digesttype' correctly 2015-01-29 14:04:21 -05:00
David Goulet
80bed1ac96 Remove obsolete workaround in dirserv_thinks_router_is_hs_dir()
Fixes #14202

Signed-off-by: David Goulet <dgoulet@ev0ke.net>
2015-01-29 12:55:19 -05:00
David Goulet
ebc59092bc Make hidden service use the config unix prefix
Signed-off-by: David Goulet <dgoulet@ev0ke.net>
2015-01-28 18:01:53 -05:00
David Goulet
bf3fb55c47 Support unix: prefix in port configuration
It's now possible to use SocksPort or any other kind of port that can use a
Unix socket like so:

  SocksPort unix:/foo/bar/unix.sock

Fixes #14451

Signed-off-by: David Goulet <dgoulet@ev0ke.net>
2015-01-28 17:55:38 -05:00
Nick Mathewson
a3de2dfde6 Merge branch 'bug11485_026_v2_squashed' 2015-01-28 14:32:19 -05:00
David Goulet
fb523b543a fixup! Refactor the use of ifdef HAVE_SYS_UN_H
Signed-off-by: David Goulet <dgoulet@ev0ke.net>
2015-01-28 14:30:23 -05:00
Andrea Shepard
bce824a9ad Actually make connections to HSes on AF_UNIX sockets 2015-01-28 14:30:23 -05:00
Andrea Shepard
ca5ba2956b Support connection_exit_connect() to AF_UNIX sockets 2015-01-28 14:30:23 -05:00
Andrea Shepard
6564291601 Handle config options for AF_UNIX hidden services rendservice.c 2015-01-28 14:30:23 -05:00
Arthur Edelstein
cb714d896c Bug #8405: Report SOCKS username/password in CIRC status events
Introduces two new circuit status name-value parameters: SOCKS_USERNAME
and SOCKS_PASSWORD. Values are enclosing in quotes and unusual characters
are escaped.

Example:

    650 CIRC 5 EXTENDED [...] SOCKS_USERNAME="my_username" SOCKS_PASSWORD="my_password"
2015-01-28 12:02:15 -05:00
Nick Mathewson
32dad3b83b Add GETINFO consensus/{valid-{after,until},fresh-until} 2015-01-28 11:28:21 -05:00
Nick Mathewson
c240eea0df more typo fixes from mcs and gk 2015-01-28 11:25:37 -05:00
Nick Mathewson
f75ca04520 Tweak tor-resolve docs and logs
Resolves 14325
2015-01-28 10:11:08 -05:00
Nick Mathewson
e9caa8645e Try to work around changes in openssl 1.1.0
Prefer not to use a couple of deprecated functions; include more
headers in tortls.c

This is part of  ticket 14188.
2015-01-28 10:00:58 -05:00
Nick Mathewson
20d0b1a04e Bump a client authorization message from debug to info.
A user wants this for 14015, and it seems fairly reasonable.
2015-01-28 09:42:28 -05:00
Nick Mathewson
9c4328c038 New GETINFO consensus/packages to expose package information from consensus 2015-01-27 16:40:32 -05:00
Nick Mathewson
1e61b45251 Fixes on prop227 comments, based on comments by mcs on #10395 2015-01-27 16:31:48 -05:00
Andrea Shepard
03d6a31716 Groundwork for AF_UNIX hidden services in rendservice.c 2015-01-27 06:22:37 +00:00
Nick Mathewson
a598d0f575 Bail early in cpuworker_onion_handshake_replyfn if the circuit is marked 2015-01-26 10:19:07 -05:00
Nick Mathewson
8f9fb3e8fa Try to fix some more memory leaks in the unit tests 2015-01-23 11:35:05 -05:00
Nick Mathewson
614fbf1812 Fix some memory leaks in new address tests 2015-01-23 11:26:40 -05:00
Nick Mathewson
034e2788f8 whitespace fixes 2015-01-23 11:18:28 -05:00
Nick Mathewson
7322de15dc Split the slow unit tests into their own binary
This can run in parallel with the faster ones and the other tests.
2015-01-23 11:15:53 -05:00
Nick Mathewson
420037dcef Merge branch 'if_addr_refactoring_squashed'
Conflicts:
	src/test/include.am
	src/test/test.c
2015-01-23 10:13:37 -05:00
rl1987
3966145dff Refactor code that looks up addresses from interfaces
Now the code has separate implementation and examination functions,
uses smartlists sanely, and has relatively decent test coverage.
2015-01-23 10:07:17 -05:00
Nick Mathewson
6c443e987d Tweak the 9969 fix a little
If we have busy nodes and excluded nodes, then don't retry with the
excluded ones enabled.  Instead, wait for the busy ones to be nonbusy.
2015-01-23 09:37:08 -05:00
Nick Mathewson
5d4bb6f61f Merge remote-tracking branch 'public/ticket9969'
Conflicts:
	src/or/directory.c
	src/or/routerlist.c
	src/or/routerlist.h
	src/test/include.am
	src/test/test.c
2015-01-23 09:36:00 -05:00
Nick Mathewson
b677ccd3ab Merge remote-tracking branch 'public/ticket13762' 2015-01-23 08:55:31 -05:00
Nick Mathewson
d8517fe843 Merge remote-tracking branch 'origin/maint-0.2.5' 2015-01-23 08:53:21 -05:00
Nick Mathewson
7cbdec578b Merge remote-tracking branch 'origin/maint-0.2.4' into maint-0.2.5 2015-01-23 08:52:55 -05:00
Nick Mathewson
df4c484021 Merge remote-tracking branch 'karsten/geoip6-jan2015' into maint-0.2.4 2015-01-23 08:52:35 -05:00
Nick Mathewson
dbd5a9a8f9 Merge remote-tracking branch 'origin/maint-0.2.3' into maint-0.2.4 2015-01-23 08:52:20 -05:00
Nick Mathewson
4a6b43bf76 fix some warnings in compat_threads.c 2015-01-22 14:22:39 -05:00
Nick Mathewson
1fb9979eb8 Move a redundant _GNU_SOURCE to where it is not redundant 2015-01-22 14:18:10 -05:00
Karsten Loesing
a9ce0cd659 Update geoip6 to the January 7 2015 database. 2015-01-22 09:58:29 +01:00
Karsten Loesing
c3f8f5ab0e Update geoip to the January 7 2015 database. 2015-01-22 09:56:54 +01:00
Nick Mathewson
38b3f9a619 use the correct free fn. spotted by dgoulet 2015-01-21 14:54:38 -05:00
Nick Mathewson
23fc1691b6 Merge branch 'better_workqueue_v3_squashed' 2015-01-21 14:47:16 -05:00
David Goulet
84f5cb749d Fix: remove whitespace and update a comment in cpuworker.c
Signed-off-by: David Goulet <dgoulet@ev0ke.net>
2015-01-21 14:31:02 -05:00
David Goulet
f52ac5be74 Fix: change copyright year in workqueue and thread tests
Signed-off-by: David Goulet <dgoulet@ev0ke.net>
2015-01-21 14:31:02 -05:00
David Goulet
d684dbb0c7 Support monotonic time for pthread_cond_timedwait
This is to avoid that the pthread_cond_timedwait() is not affected by time
adjustment which could make the waiting period very long or very short which
is not what we want in any cases.

Signed-off-by: David Goulet <dgoulet@ev0ke.net>
2015-01-21 14:29:12 -05:00
Nick Mathewson
f0415c1600 Merge branch 'bug9819' 2015-01-21 13:00:26 -05:00
Nick Mathewson
523e920d53 fix a comment 2015-01-21 13:00:14 -05:00
Nick Mathewson
3c8dabf69a Fix up some workqueue/threading issues spotted by dgoulet. 2015-01-21 12:26:41 -05:00
Nick Mathewson
ac5b70c700 handle EINTR in compat_*threads.c 2015-01-21 12:18:11 -05:00
Nick Mathewson
e7e33d4b04 Merge branch 'bug14084' 2015-01-20 14:07:37 -05:00
Nick Mathewson
9ddc1fb10c Merge remote-tracking branch 'dgoulet/bug14224_025_v1' 2015-01-20 14:02:07 -05:00
Nick Mathewson
da423532f7 Merge branch 'ticket14254_squashed' 2015-01-20 13:46:56 -05:00
Nick Mathewson
78c53eff85 Fix SocksSocket 0. That was easy! 2015-01-20 13:46:44 -05:00
Nick Mathewson
061682c829 Some days I just can't C. 2015-01-19 11:58:40 -05:00
Nick Mathewson
2e8b8c8698 Make check-spaces happier. 2015-01-19 11:56:03 -05:00
Nick Mathewson
f92b01b963 Remove a now-needless testing workaround. 2015-01-19 11:55:52 -05:00
Nick Mathewson
a8dd930274 Replace a 4 with a 6; fix a bug that nobody noticed :/
Fixes 14280 bugfix on 1053af0b9c in 0.2.4.7-alpha.
2015-01-19 11:51:08 -05:00
Nick Mathewson
1053af0b9c Merge branch 'bug7555_v2_squashed'
Conflicts:
	src/or/connection_edge.c
2015-01-19 11:43:41 -05:00
Nick Mathewson
758d77130c Add a bunch of new comments to explain connection_ap_rewrite{,_and_attach}
Also, do a little light refactoring to move some variable declarations
around and make a few things const

Also fix an obnoxious bug on checking for the DONE stream end reason.
It's not a flag; it's a possible value or a variable that needs to be
masked.
2015-01-19 11:30:22 -05:00
David Goulet
b5525476f5 Fix: close intro circuit if no more intro points are usable
Once a NACK is received on the intro circuit, tor tries an other usable one
by extending the current circuit to it. If no more intro points are usable,
now close the circuit. Also, it's reason is changed before closing it so we
don't report again an intro point failure and trigger an extra HS fetch.

Fixes #14224

Signed-off-by: David Goulet <dgoulet@ev0ke.net>
2015-01-19 10:23:58 -05:00
David Goulet
b3c1152bae Fix: close intro circuit if no more intro points are usable
Once a NACK is received on the intro circuit, tor tries an other usable one
by extending the current circuit to it. If no more intro points are usable,
now close the circuit.

Fixes #14224

Signed-off-by: David Goulet <dgoulet@ev0ke.net>
2015-01-19 09:44:11 -05:00
Nick Mathewson
63765399eb Merge remote-tracking branch 'public/ticket13037'
Conflicts:
	src/or/config.c
2015-01-18 16:07:08 -05:00
Roger Dingledine
5aa55a1369 Remove the unused rend_cache_entry_t->received field.
(Patch from arma, commit message from nick.)

Closes #14222.
2015-01-18 15:53:06 -05:00
Nick Mathewson
fae72a8d0a Merge remote-tracking branch 'public/bug14219_025' 2015-01-18 15:41:13 -05:00
Roger Dingledine
9407040c59 Do not replace a HS descriptor with a different replica of itself
This fixes a bug where we'd fetch different replicas of the same
descriptor for a down hidden service over and over, until we got lucky
and fetched the same replica twice in a row.

Fixes bug 14219; bugfix on 0.2.0.10-alpha.

(Patch from Roger; commit message and changes file by Nick.)
2015-01-18 15:39:12 -05:00
Nick Mathewson
efdac2a68c Merge remote-tracking branch 'public/bug14261_025'
Conflicts:
	src/or/directory.c
2015-01-18 15:28:35 -05:00
Nick Mathewson
ceb6dee465 Increase limit for status vote download size by a factor of 5.
We've started to hit the limit here.  We introduced the limit in
0.1.2.5-alpha.  This fixes bug 14261, but we should have a smarter way
to not actually do the behavior this permits.  See #14267 for a ticket
about fixing that.
2015-01-18 15:25:29 -05:00
Nick Mathewson
54e4aaf52c Fix memory leak in connection_ap_handshake_rewrite_and_attach()
Spotted by asn.  #14259.  Bugfix on 368eb6a97 in 0.2.0.1-alpha.
2015-01-18 14:19:26 -05:00
Nick Mathewson
79e12da861 Merge remote-tracking branch 'public/bug12485' 2015-01-18 13:49:30 -05:00
Nick Mathewson
ba17cdfb0a fix another mingw64 unit test warnin 2015-01-16 11:49:58 -05:00
Nick Mathewson
31838bd783 changes suggested by weasel 2015-01-16 11:46:20 -05:00
Nick Mathewson
485fdcf826 Unify parse_unix_socket_config and parse_port_config
This incidentally makes unix SocksSocket support all the same options
as SocksPort.

This patch breaks 'SocksSocket 0'; next will restore it.

Resolves 14254.
2015-01-16 11:35:48 -05:00
Nick Mathewson
bbad23bf37 No, client-side DNS cacheing should not be on by default. 2015-01-16 09:32:22 -05:00
Nick Mathewson
4b23b398a3 Merge branch 'bug8546_squashed'
Conflicts:
	src/or/connection.c
	src/or/or.h
	src/or/relay.c
2015-01-16 09:31:50 -05:00
Nick Mathewson
49bdfbabb4 Replace field-by-field copy with memcpy for entry_port_cfg 2015-01-16 09:23:03 -05:00
Nick Mathewson
13dac5e463 Move entry_port_cfg_t fields in entry_connection_t
Also rename some options for uniformity, and apply this script:

@@
entry_connection_t *conn;
@@
 conn->
+entry_cfg.
\(
 isolation_flags
\|
 session_group
\|
 socks_prefer_no_auth
\|
 ipv4_traffic
\|
 ipv6_traffic
\|
 prefer_ipv6
\|
 cache_ipv4_answers
\|
 cache_ipv6_answers
\|
 use_cached_ipv4_answers
\|
 use_cached_ipv6_answers
\|
 prefer_ipv6_virtaddr
\)
2015-01-16 09:22:58 -05:00
Nick Mathewson
58d17add5e Combine entry_port_cfg_t fields in listener_connection_t
Also, revise the code using these options with this cocci script:

@@
listener_connection_t *conn;
@@
 conn->
+entry_cfg.
\(
 isolation_flags
\|
 session_group
\|
 socks_prefer_no_auth
\|
 ipv4_traffic
\|
 ipv6_traffic
\|
 prefer_ipv6
\|
 cache_ipv4_answers
\|
 cache_ipv6_answers
\|
 use_cached_ipv4_answers
\|
 use_cached_ipv6_answers
\|
 prefer_ipv6_virtaddr
\)
2015-01-16 09:22:53 -05:00
Nick Mathewson
f444f2b1d3 Split client-specific and server-specific parts of port_cfg_t
Also, apply this cocci script to transform accesses. (Plus manual
migration for accesses inside smartlist_foreach loops.)

@@

port_cfg_t *cfgx;
@@
 cfgx->
+server_cfg.
\(
 no_advertise
\|
 no_listen
\|
 all_addrs
\|
 bind_ipv4_only
\|
 bind_ipv6_only
\)

@@

port_cfg_t *cfgx;
@@
 cfgx->
+entry_cfg.
\(
 isolation_flags
\|
 session_group
\|
 socks_prefer_no_auth
\|
 ipv4_traffic
\|
 ipv6_traffic
\|
 prefer_ipv6
\|
 cache_ipv4_answers
\|
 cache_ipv6_answers
\|
 use_cached_ipv4_answers
\|
 use_cached_ipv6_answers
\|
 prefer_ipv6_virtaddr
\)
2015-01-16 09:22:49 -05:00
Nick Mathewson
2329d9fe37 Fix a compilation warning in the unit tests 2015-01-15 12:56:57 -05:00
Nick Mathewson
746bb55851 Ignore warning for redundant decl in openssl/srtp.h
Backports some commits from tor master.
2015-01-15 12:38:08 -05:00
Nick Mathewson
3368b0c9f2 Add string for IP_NOW_REDUNDANT in circuit_end_reason_to_control_string
Closes 14207; bugfix on 0.2.6.2-alpha.
2015-01-15 11:53:20 -05:00
George Kadianakis
bb56fd1f8f unittests: use tt_i64_op() instead of tt_int_op() in laplace tests.
Do this in even more places to try to fix the build.
2015-01-15 11:08:22 -05:00
Nick Mathewson
a52e549124 Update workqueue implementation to use a single queue for the work
Previously I used one queue per worker; now I use one queue for
everyone.  The "broadcast" code is gone, replaced with an idempotent
'update' operation.
2015-01-15 11:05:22 -05:00
Roger Dingledine
660a35d97c fix typo 2015-01-15 11:03:13 -05:00
Nick Mathewson
52d6cb03a7 Merge remote-tracking branch 'andrea/bug12194_alt' 2015-01-15 10:26:35 -05:00
Nick Mathewson
88e36eaf0e Fix the checkdir/perms test when umask==077
Fixes 14215; bugfix on 0.2.6.2-alpha.  Reported by "cypherpunks".
2015-01-15 10:24:27 -05:00
Andrea Shepard
f7bb60e202 Add comments in rendclient.c noting that certain functions involved in handling ended HS connection attempts must be idempotent 2015-01-15 15:19:31 +00:00
George Kadianakis
354ddf8712 unittests: use tt_i64_op() instead of tt_int_op() in laplace tests. 2015-01-15 15:14:33 +00:00
George Kadianakis
25e7821bb1 unittests: Change some tt_assert()s to tt_int_op()s. 2015-01-15 14:43:58 +00:00
George Kadianakis
45bc5a0743 Restrict sample values of the Laplace distribution to int64_t.
This helps avoid undefined behavior from casting big double values to
int64_t. Fixes #14090.
2015-01-15 14:43:58 +00:00
Nick Mathewson
3668a4126e Merge remote-tracking branch 'public/bug13397' 2015-01-14 14:15:29 -05:00
Nick Mathewson
1686f81ac2 Merge remote-tracking branch 'origin/maint-0.2.5' 2015-01-14 13:30:10 -05:00
Nick Mathewson
051ad788e0 Incorporate some comments based on notes from dgoulet 2015-01-14 11:31:14 -05:00
Nick Mathewson
fb5ebfb507 Avoid chan/circ linear lookups for requests
The solution I took is to not free a circuit with a pending
uncancellable work item, but rather to set its magic number to a
sentinel value.  When we get a work item, we check whether the circuit
has that magic sentinel, and if so, we free it rather than processing
the reply.
2015-01-14 11:31:12 -05:00
Nick Mathewson
6c9c54e7fa Remove if (1) indentation in cpuworker.c
To avoid having diffs turn out too big, I had replaced some unneeded
ifs and fors with if (1), so that the indentation would still work out
right.  Now I might as well clean those up.
2015-01-14 11:28:26 -05:00
Nick Mathewson
1e896214e7 Refactor cpuworker to use workqueue/threadpool code. 2015-01-14 11:23:34 -05:00
Nick Mathewson
cc6529e9bb Fix check-spaces 2015-01-14 11:19:35 -05:00
Nick Mathewson
e5f8c772f4 Test and fix workqueue_entry_cancel(). 2015-01-14 11:17:46 -05:00
Nick Mathewson
ebbc177005 Add shutdown and broadcast support to test_workqueue. 2015-01-14 11:17:46 -05:00
Nick Mathewson
81354b081b Add unit test for thread IDs. 2015-01-14 11:17:43 -05:00
Nick Mathewson
7a63005220 Basic unit test for condition variables. 2015-01-14 11:17:09 -05:00
Nick Mathewson
e47a90a976 "Recursive" locks, not "reentrant" locks. Duh. 2015-01-14 11:12:40 -05:00
Nick Mathewson
d69717f61b Use correct (absolute) time for pthread_cond_timedwait 2015-01-14 11:09:52 -05:00
Nick Mathewson
9fdc0d0594 Fix windows compilation of condition code 2015-01-14 11:09:51 -05:00
Nick Mathewson
d850ec8574 Fix linux compilation (pipe2 needs _GNU_SOURCE) 2015-01-14 11:09:51 -05:00
Nick Mathewson
74b782645a Move thread tests into their own module 2015-01-14 11:09:47 -05:00
Nick Mathewson
c51f7c23e3 Test a little more of compat_threads.c 2015-01-14 11:05:56 -05:00
Nick Mathewson
3868b5d210 Rename mutex_for_cond -> mutex_nonreentrant
We'll want to use these for other stuff too.
2015-01-14 11:05:56 -05:00
Nick Mathewson
93ad89e9d2 Rename bench_workqueue -> test_workqueue and make it a unit test. 2015-01-14 11:05:56 -05:00
Nick Mathewson
b2db3fb462 Documentation for new workqueue and condition and locking stuff 2015-01-14 11:05:54 -05:00
Nick Mathewson
4abbf13f99 Add a way to tell all threads to do something. 2015-01-14 11:01:21 -05:00
Nick Mathewson
51bc0e7f3d Isolate the "socketpair or a pipe" logic for alerting main thread
This way we can use the linux eventfd extension where available.
Using EVFILT_USER on the BSDs will be a teeny bit trickier, and will
require libevent hacking.
2015-01-14 11:01:19 -05:00
Nick Mathewson
c7eebe237d Make pending work cancellable. 2015-01-14 10:56:27 -05:00
Nick Mathewson
a82604b526 Initial workqueue implemention, with a simple test.
It seems to be working, but more tuning is needed.
2015-01-14 10:56:24 -05:00
Nick Mathewson
6c9363310a Specialize handling for mutexes allocated for condition variables
(These must not be reentrant mutexes with pthreads.)
2015-01-14 10:52:56 -05:00
Nick Mathewson
65016304d2 Add tor_cond_init/uninit 2015-01-14 10:49:59 -05:00
Nick Mathewson
e865248156 Add a timeout to tor_cond_wait; add tor_cond impl from libevent
The windows code may need some tweaks for it to compile; I've not
tested it yet.
2015-01-14 10:47:39 -05:00