|
1c3147f5
|
2013-04-10T18:03:16
|
|
Add a test with an active_later event at event_base_free time.
|
|
|
2fad0f3d
|
2013-04-05T15:06:54
|
|
Add an environment variable (EVENT_DEBUG_MODE) to run unit tests in debug mode
Not all tests currently pass with debug mode on.
|
|
|
f935e215
|
2013-03-15T09:33:28
|
|
build test/test-script.sh on systems with a less-featureful $<
|
|
|
787fd748
|
2013-03-15T09:33:13
|
|
Make --disable-libevent-regress work again
|
|
13676535
|
2013-03-01T12:01:42
|
|
Test: decoding just part of string with evhttp_decode_uri_internal()
|
|
c9ad3af2
|
2013-02-25T20:13:01
|
|
test filling up the hash table a bit
|
|
|
2078e9b4
|
2013-02-14T20:14:37
|
|
make sure caching works, and we don't leak memory
|
|
|
72965126
|
2013-02-07T17:06:49
|
|
a program to print out the error strings for winsock errors
|
|
|
97094614
|
2013-02-14T14:13:11
|
|
Merge remote-tracking branch 'origin/patches-2.0'
|
|
b6182042
|
2013-02-14T09:54:56
|
|
fix #73 and fix http_connection_fail_test to catch it
|
|
|
96150dd0
|
2013-02-13T11:38:57
|
|
Unit test for event_remove_timer with EV_PERSIST.
Patch from dcicppin on sourceforge.
|
|
|
74e52db4
|
2013-02-11T11:25:25
|
|
Merge remote-tracking branch 'ppelleti/nmake-clean-exes'
|
|
|
2863c837
|
2013-02-08T22:10:05
|
|
Avoid using $(top_srcdir) in TESTS.
Newer automakes don't like this.
|
|
|
a55514ee
|
2013-02-08T22:08:18
|
|
Avoid using top_srcdir in TESTS-new automakes do not like this
|
|
|
974bfa07
|
2013-02-07T17:14:18
|
|
remove all exes on "make clean", not just regress.exe
|
|
2ecd8947
|
2013-02-04T13:49:08
|
|
Merge pull request #39 from azat/fix-http-for-ipv6
Fix ipv6 support for http. When URL contain domain, not IP address.
|
|
0dda56a4
|
2013-02-03T19:08:37
|
|
Preliminary changes for Minix3.
|
|
|
9ec88bdf
|
2013-01-27T02:17:02
|
|
Add regress test ipv6_for_domain.
Written for commit 71e709c7829275a594f767b27468b1b52e8b5bb9.
Fix ipv6 support for http. When URL contain domain, not IP address.
|
|
|
c1420690
|
2012-11-19T09:07:13
|
|
Fix a bug when running "make verify" out-of-tree
|
|
|
173d7823
|
2012-11-18T01:40:33
|
|
Fix a gcc warning in rtrim test
|
|
|
f38e0788
|
2012-11-18T01:36:03
|
|
Make ssl version check in unit tests work
|
|
|
e3b2e086
|
2012-11-16T16:15:03
|
|
Add an event_remove_timer() to remove timer on an event without deleting it
|
|
|
5a9a0141
|
2012-11-16T16:17:07
|
|
Fix a couple of compile warnings in the unit tests
|
|
|
c6ff3812
|
2012-11-16T11:38:53
|
|
Remove internal ws from multiline http headers correctly
According to RFC2616:
All linear white space, including folding, has the same semantics
as SP. A recipient MAY replace any linear white space with a single
SP before interpreting the field value or forwarding the message
downstream.
|
|
|
ac425197
|
2012-11-16T11:29:34
|
|
Remove trailing tabs in HTTP headers as well.
|
|
|
aa59d805
|
2012-11-16T11:13:29
|
|
Clean up rtrim implementation
If I understand the C standard correctly, you can't actually point
at a position immediately _before_ the start of an object; only at the
position immediately after.
According to J.2 in the standard, in its big list of undefined behavior:
"The behavior is undefined in the following circumstances:
...
— Addition or subtraction of a pointer into, or just beyond, an
array object and an integer type produces a result that does not
point into, or just beyond, the same array object (6.5.6)."
So we've got to fix rtrim to not do that. Also, make it unit tested,
and give it an evutil_*_ name.
|
|
|
efb644ed
|
2012-11-16T10:29:37
|
|
Merge remote-tracking branch 'origin/patches-2.0'
Conflicts:
test/rpcgen_wrapper.sh
|
|
|
f3009e48
|
2012-11-16T10:28:19
|
|
Make rpcgen_wrapper.sh work on systems without a "python2" binary
|
|
|
4343edf3
|
2012-11-16T10:17:34
|
|
Add a unit test in which an event is created with event_base_once() but never fires
|
|
|
1f5a48d1
|
2012-11-15T11:45:12
|
|
Merge remote-tracking branch 'origin/patches-2.0'
|
|
|
ac009f92
|
2012-11-15T11:43:20
|
|
Warn when openssl version in unit test mismatches compiled version.
|
|
|
c2f30863
|
2012-11-15T11:42:14
|
|
Fix renegotiation test to work around openssl 1.0.1 bug
There's a bug in openssl 1.0.1 where TLS1.1 and TLS1.2 can't
renegotiate with themselves. When testing renegotiation with OpenSSL
>=1.0.1 and <1.0.1d, disable those protocols.
|
|
|
1d9d5110
|
2012-11-15T11:40:33
|
|
Use the same CFLAGS for openssl when building unit tests as with libevent
|
|
e9f8feba
|
2012-11-06T17:37:28
|
|
Add a new callback to get called on evbuffer_file_segment free
|
|
|
64051b90
|
2012-11-06T17:42:46
|
|
Clean up unit test for evbuffer_file_segment_add_cleanup_cb a little
|
|
|
56e48c10
|
2012-11-02T10:58:02
|
|
Fix a few mingw64 incompatibilities introduced since 2.0
|
|
|
2e6a9850
|
2012-11-01T18:12:07
|
|
Merge remote-tracking branch 'github/20_win64_compilation' into 21_win64_compilation
Conflicts:
event.c
http.c
sample/event-read-fifo.c
test/regress_bufferevent.c
|
|
|
94866c27
|
2012-11-01T17:38:34
|
|
Compile without warnings on mingw64
This is mostly a matter of catching cases where we were still
assuming that evutil_socket_t could be used as an int.
|
|
4b5f32a5
|
2012-10-23T13:26:44
|
|
Use python2 rather than python
python may refer to either python2 or python3 so rather by explicit by
using python2.
See PEP 394 - http://www.python.org/dev/peps/pep-0394/ for more
details.
|
|
|
1e3123da
|
2012-08-28T16:15:14
|
|
Fix handling of no-python case for nonrecursive make
|
|
|
607a8ff9
|
2012-08-28T16:14:38
|
|
Be quiet when making regress.gen.[ch]
|
|
|
371a1237
|
2012-08-28T16:07:56
|
|
Make quiet build even quieter
|
|
|
6cdfeebe
|
2012-08-28T16:01:14
|
|
Rename subordinate Makefile.ams to include.am
|
|
|
7092f3b6
|
2012-08-28T15:58:18
|
|
Switch to non-recursive makefiles
This approach should let Make work better, let libevent build faster
(especially when using a parallelized build), and basically make the
Makefiles easier to maintain.
See http://miller.emu.id.au/pmiller/books/rmch/ for more info on why
you'd want to do this.
This is due to an idea from Zack Weinberg; the patch is my own.
|
|
|
07e132e3
|
2012-08-23T10:46:29
|
|
Rename event_enable_lock_debuging() to ..._debugging()
Keep the misspelled version around for backward compatibility
Based on a patch by Diwaker Gupta.
|
|
|
f1bd9388
|
2012-08-02T15:02:27
|
|
Avoid test -e; older shs don't have one.
Reported by Paul Croome.
Report forwarded by Dave Hart.
|
|
|
34c8f316
|
2012-08-02T12:36:05
|
|
Add deferred_cb_skew to list of timing-dependent tests
|
|
|
403b0d6e
|
2012-07-31T12:03:48
|
|
Merge remote-tracking branch 'ppelleti/windows-ssl'
|
|
|
9852107f
|
2012-07-26T10:43:13
|
|
Merge remote-tracking branch 'origin/patches-2.0'
Conflicts:
buffer.c
http.c
|
|
|
a2006c00
|
2012-07-26T10:37:47
|
|
Move assignment outside tt_assert in ssl unit tests. Appeases coverity.
|
|
|
b9e73297
|
2012-07-26T10:37:13
|
|
Add checks to various return values in unit tests. Found by coverity
|
|
|
6355b2a4
|
2012-07-26T10:16:47
|
|
Remove unused variable; spotted by coverity
|
|
|
225344cf
|
2012-07-26T10:11:17
|
|
Fix a shadowed variable in addfile_test_readcb; found by coverity
|
|
|
92817a11
|
2012-07-26T10:11:00
|
|
Avoid memory leak in test_event_calloc unit test; found by coverity
|
|
|
aa501e1a
|
2012-07-26T10:10:35
|
|
Add checks for functions in test-ratelim.c; found by Coverity
|
|
|
7ef04e4b
|
2012-07-26T10:09:13
|
|
Merge remote-tracking branch 'origin/patches-2.0'
Conflicts:
configure.in
test/regress_buffer.c
test/regress_dns.c
|
|
|
3cde5bf5
|
2012-07-26T10:04:43
|
|
Avoid more crashes/bad calls in unit tests; found by coverity
|
|
|
f021c3d7
|
2012-07-26T09:59:47
|
|
Add some missing null checks to unit tests; found by coverity
|
|
|
270f279f
|
2012-07-26T09:53:20
|
|
Fix resource leaks in the unit tests; found by coverity
|
|
|
4f3732d7
|
2012-07-26T09:35:43
|
|
Fix various check-after-dereference issues in unit tests: found by coverity
|
|
|
f1ce15d9
|
2012-07-25T16:18:33
|
|
Set umask before calling mkstemp in unit tests. Found by coverity
|
|
|
13525c50
|
2012-07-25T16:07:22
|
|
Fix another possible uninitialized read in dns regression tests. Found by coverity.
|
|
|
6fdbf3a2
|
2012-07-25T16:06:51
|
|
Merge remote-tracking branch 'origin/patches-2.0'
|
|
|
2259777c
|
2012-07-25T16:05:36
|
|
Fix possible uninitialized read in dns regression tests. Found by coverity.
|
|
|
55533467
|
2012-07-25T15:59:50
|
|
Remove deadcode in http regression tests. Found by coverity.
|
|
|
52a0039b
|
2012-07-20T15:36:15
|
|
Fix tinytset_skip to work with new tinytest_set_flag_ signature
|
|
|
eba0eb2e
|
2012-06-08T19:47:23
|
|
attempt to support OpenSSL in Makefile.nmake
|
|
|
2b6fe8bb
|
2012-06-29T12:47:51
|
|
Make load-dependent monotonic timer tests off-by-default
You can reenable them with @all, or say @fine_timing just to get
them.
|
|
|
56710337
|
2012-06-29T12:47:39
|
|
Avoid (unlikely) overflow in bench_httpclient.c
|
|
|
c8009d29
|
2012-06-29T12:47:03
|
|
Check return value of write() in regress.c
|
|
|
f2cea873
|
2012-06-28T13:18:28
|
|
Make all tests that hit the network disabled by default
To run only these tests, run test/regress with @live_net. To run all
tests, run test/regress with @all.
|
|
|
ef7c4f79
|
2012-06-28T12:58:46
|
|
Update to latest tinytest (911b4f0349377)
|
|
|
def3b834
|
2012-06-28T12:57:58
|
|
Make test_evutil_monotonic a little more tolerant
|
|
|
108896ab
|
2012-06-28T11:42:25
|
|
Merge branch 'master' of ssh://github.com/libevent/libevent
|
|
|
e2d7d98d
|
2012-06-28T11:37:45
|
|
remove a now-unused var in thread_deferred_cb_skew test
|
|
|
8a739b34
|
2012-06-28T11:36:52
|
|
Merge branch '21_event_callback_v3'
|
|
|
ea92fba3
|
2012-06-25T13:55:45
|
|
Avoid a resource leak on error in http client benchmark
|
|
|
90daab61
|
2012-06-06T17:06:57
|
|
Merge remote-tracking branch 'origin/patches-2.0'
|
|
|
3b2529ac
|
2012-06-06T21:57:12
|
|
Fix a couple of memory leaks (found with Valgrind).
|
|
|
3881d8fd
|
2012-06-03T15:19:53
|
|
Fix a comment in test-fdleak after 077c7e949.
|
|
|
c0e425ab
|
2012-05-09T11:06:06
|
|
Restore our priority-inversion-prevention code with deferreds
Back when deferred_cb stuff had its own queue, the queue was always
executed, but we never ran more than 16 callbacks per iteration.
That made for two problems:
1: Because deferred_cb stuff would always run, and had no priority,
it could cause priority inversion.
2: It doesn't respect the max_dispatch_interval code.
Then, when I refactored deferred_cb to be a special case of
event_callback, that solved the above issues, but made for two more
issues:
3: Because deferred_cb stuff would always get the default priority,
it could could low-priority bufferevents to get too much priority.
4: With code like bufferevent_pair, it's easy to get into a
situation where two deferreds keep adding one another, preventing
the event loop from ever actually scanning for more events.
This commit fixes the above by giving deferreds a better notion of
priorities, and by limiting the number of deferreds that can be
added to the _current_ loop iteration's active queues. (Extra
deferreds are put into the active_later state.)
That isn't an all-purpose priority inversion solution, of course: for
that, you may need to mess around with max_dispatch_interval.
|
|
|
a4079aa8
|
2012-04-06T11:05:35
|
|
Replace more deferred_cb names with event_callback
|
|
|
ae2b84b2
|
2012-04-06T04:33:19
|
|
Replace deferred_cbs with event_callback-based implementation.
|
|
|
745a63db
|
2012-04-06T03:00:40
|
|
Add "active later" event_callbacks to supersede deferred
An event or event callback can now be in an additional state: "active
later". When an event is in this state, it will become active the
next time we run through the event loop. This lets us do what we
wanted to with deferred callbacks: make a type of active thing that
avoids infinite circular regress in a way that starves other events or
exhausts the stack. It improves on deferred callbacks by respecting
priorities, and by having a non-kludgy way to avoid event starvation.
|
|
|
e0aa0866
|
2012-05-03T12:21:57
|
|
Merge remote-tracking branch 'github/patches-2.0'
Conflicts:
test/regress.c
|
|
|
e06206eb
|
2012-05-03T12:15:11
|
|
Oops; fix the *right* windows compile issue
|
|
|
be7a0be8
|
2012-05-03T12:05:52
|
|
Fix win32 build issues: sleep, pid_t
|
|
|
a1630260
|
2012-05-01T13:22:56
|
|
Merge branch '21_robust_monotonic'
|
|
|
2e882a07
|
2012-04-30T22:00:05
|
|
Merge remote-tracking branch 'origin/patches-2.0'
Conflicts:
event.c
|
|
|
30f06715
|
2012-04-30T21:45:43
|
|
Merge remote-tracking branch 'github/20_active_prio_inv' into patches-2.0
|
|
|
b3887cdf
|
2012-04-11T21:33:27
|
|
Work-around a stupid gcov-breaking bug in OSX 10.6
This only affects the unit tests.
Fix found at http://rachelbythebay.com/w/2011/07/12/forkcrash/
(Backport from 2.1)
|
|
|
2bfda401
|
2012-04-30T17:30:48
|
|
If a higher-priority event becomes active, don't continue running events of the current priority.
Bug found by Ralph Castain.
|
|
|
26c75828
|
2012-04-26T16:22:03
|
|
When PRECISE_TIMERS is set with epoll, use timerfd for microsecond precision
The epoll interface ordinarily gives us one-millisecond
precision, so on Linux it makes perfect sense to use the
CLOCK_MONOTONIC_COARSE timer. But when the user has set the new
PRECISE_TIMER flag for an event_base (either by the
EVENT_BASE_FLAG_PRECISE_TIMER flag, or by the EVENT_PRECISE_TIMER
environment variable), they presumably want finer granularity.
On not-too-old Linuxes, we can achieve this using the Timerfd
mechanism, which accepts nanosecond granularity and understands
posix clocks. It's a little more expensive than just calling
epoll_wait(), so we won't do it by default.
|
|
|
7428c78a
|
2012-04-26T14:35:18
|
|
Improve the monotonic-time unit test: make it check the step size
|
|
|
630f077c
|
2012-04-26T11:56:59
|
|
Simple unit tests for monotonic timers
|
|
|
71bca50f
|
2012-04-20T12:27:12
|
|
Split out time-related prototypes into time-internal.h
|
|
|
2711cda3
|
2012-04-20T12:14:20
|
|
Split long lists in Makefile.am into one-item-per-line
|
|
|
21205b83
|
2012-04-20T11:53:32
|
|
Shave 700 msec off the persistent_timeout_jump test
|
|
|
d992d911
|
2012-04-20T11:51:33
|
|
Merge remote-tracking branch 'origin/patches-2.0'
|
|
|
dfd808cb
|
2012-04-19T00:25:12
|
|
If time has jumped so we'd reschedule a periodic event in the past, schedule it for the future instead
Fixes an issue reported on libevent-users in the thread "a dead
looping bug when changing system time backward". Previously, if time
jumped forward 1 hour[*] and we had a one-second periodic timer event,
that event would get invoked 3600 times. That's almost certainly not
what anybody wants.
In a future version of Libevent, we should expose the amount of time
that the callbac kwould have been invoked somehow.
[*] Forward time jumps can happen with nonmonotonic clocks, or with
clocks that jump on suspend/resume. It can also happen from
Libevent's point of view if the user exits from event_base_loop() and
doesn't call it again for a while.
|