|
117dc921
|
2018-10-24T11:51:57
|
|
Filter link-local IPv4 addresses in evutil_found_ifaddr()
Fixes: #668
|
|
|
0ec12bc8
|
2018-10-23T00:12:23
|
|
Merge branch 'http-request-line-parsing'
* http-request-line-parsing:
Fix http https_basic/https_filter_basic under valgrind (increase timeout)
http: cover various non RFC3986 conformant URIs
http: allow non RFC3986 conformant during parsing request-line (http server)
http: do not try to parse request-line if we do not have enough bytes
http: allow trailing spaces (and only them) in request-line (like nginx)
http: cleanup of the request-line parsing
|
|
|
15bfe712
|
2018-10-22T23:38:42
|
|
http: cover various non RFC3986 conformant URIs
- http/basic_trailing_space -- covers cases when there is trailing space
after the request line (nginx handles this)
- http/simple_nonconformant -- covers non RFC3986 conformant URIs
|
|
|
5f1b4dfa
|
2018-10-23T00:06:47
|
|
Fix http https_basic/https_filter_basic under valgrind (increase timeout)
|
|
|
b94d913d
|
2018-10-22T23:52:46
|
|
http: allow non RFC3986 conformant during parsing request-line (http server)
Reported-by: lsdyst@163.com
|
|
|
64ead341
|
2018-10-22T23:56:50
|
|
http: do not try to parse request-line if we do not have enough bytes
|
|
|
254fbc81
|
2018-10-22T23:56:19
|
|
http: allow trailing spaces (and only them) in request-line (like nginx)
|
|
|
6cf659b0
|
2018-10-22T23:25:01
|
|
http: cleanup of the request-line parsing
|
|
|
2f43d1d4
|
2018-10-22T01:06:48
|
|
Disable parallel jobs for the osx (due to CPU time deficit) in travis-ci
As you can see right now linux workers has zero failed tests, while osx
workers has 18 failed tests:
[bufferevent_connect_hostname_emfile FAILED]
[bufferevent_pair_release_lock FAILED]
[bufferevent_timeout FAILED]
[bufferevent_timeout_filter FAILED]
[bufferevent_timeout_pair FAILED]
[common_timeout FAILED]
[del_wait FAILED]
[immediatesignal FAILED]
[loopexit FAILED]
[loopexit_multiple FAILED]
[monotonic_res FAILED]
[no_events FAILED]
[persistent_active_timeout FAILED]
[persistent_timeout_jump FAILED]
[signal_switchbase FAILED]
[signal_while_processing FAILED]
[simpletimeout FAILED]
[usleep FAILED]
And this patch should remove from this list time related failures
(though maybe not all of them).
|
|
|
5d3e8c15
|
2018-10-22T01:00:54
|
|
Use matrix.fast_finish for the travis-ci to mark the build ASAP
AFAIR there is shortage of osx workers on travis-ci, IOW builds that
requires them can wait fair amount of time in the queue by just waiting.
Plus linux workers AFAICS can run multiple jobs in parallel (4-5), while
osx does not.
Hence if we do allow failures for osx (and right now they have a lot of
failed tests) let's mark build result based on tests under linux only.
So in a nut shell this will reduce build time from 5 hours to 20-30
minutes.
|
|
|
6ac8e775
|
2018-10-21T18:31:01
|
|
Simplify bufferevent timeout tests to reduce CPU usage in between start/compare
Between start (setting "started_at") and comparing the time when
timeouts triggered with the start (test_timeval_diff_eq), there is too
much various things that can introduce extra delays and eventually could
fail the test on machine with shortage of CPU.
And this is exactly what happend on:
- travis-ci
- #262
Here is a simple reproducer that I came up with for this issue:
docker run --cpus=0.01 -e LD_LIBRARY_PATH=$PWD/lib -e PATH=/usr/bin:/bin:$PWD/bin -v $PWD:$PWD --rm -it debian:testing regress --no-fork --verbose bufferevent/bufferevent_timeout
Under limited CPU (see reproducer) the test almost always has problems
with that "write_timeout_at" exceed default timeval diff tolerance
(test_timeval_diff_eq() has 50 tolerance), i.e.:
FAIL ../test/regress_bufferevent.c:1040: assert(labs(timeval_msec_diff(((&started_at)), ((&res1.write_timeout_at))) - (100)) <= 50): 101 vs 50
But under some setup write timeout can even not triggered, and the
reason for this is that we write to the bufferevent 1024*1024 bytes, and
hence if evbuffer_write_iovec() will has some delay after writev() and
not send more then one vector at a time [1], it is pretty simple to
trigger, i.e.:
FAIL ../test/regress_bufferevent.c:1040: assert(labs(timeval_msec_diff(((&started_at)), ((&res1.write_timeout_at))) - (100)) <= 50): 1540155888478 vs 50
[1]: https://gist.github.com/azat/b72773dfe7549fed865d439e03de05c1
So this patch just send static small payload for all cases (plus a few
more asserts added).
The outcome of this patch is that all regression tests passed on
travis-ci for linux box [2]. While before it fails almost always [3].
Also reproducer with CPU limiting via docker also survive some
iterations (and strictly speaking it should has less CPU then travis-ci
workers I guess).
[2]: https://travis-ci.org/azat/libevent/builds/444391481
[3]: https://travis-ci.org/libevent/libevent/builds/444336505
|
|
3e6553a1
|
2018-08-09T13:04:52
|
|
evdns: handle NULL filename explicitly
Otherwise we will try to open NULL filename and got EFAULT and anyway
will got the same return code from evdns_base_resolv_conf_parse_impl()
Closes: #680 (cherry-picked)
|
|
|
fb090aa6
|
2018-10-21T03:18:48
|
|
Revert "travis-ci: use travis_retry to reduce number of problems with flacky tests"
That was a bad idea, because all this patch did is just uses extra CPU
time on travis-ci workers and eventually fail the build anyway.
It also increases time of executing one matrix entry (x3).
This reverts commit 7004ee8e869faabe3a284fb3d7eb859fb50ef90c.
|
|
|
bd2184d8
|
2018-10-21T03:15:34
|
|
Merge branch 'regress-dns-fixes'
* regress-dns-fixes:
Do not rely on getservbyname() for most of the dns regression tests
Turn off dns/getaddrinfo_race_gotresolve by default
Fix an error for debug locking in dns/getaddrinfo_race_gotresolve
|
|
|
2cb8eae7
|
2018-10-21T03:03:25
|
|
Do not rely on getservbyname() for most of the dns regression tests
There is only one test that uses service name getaddrinfo_async, which
manually check whether it works or not, other should not assume that it
is available and works.
There was already an attempt to overcome some possible limitations, like
lack of "http" in /etc/services in
d6bafbbeb27ff3943d6f3b6783bcded76384c31e ("test/dns: replace servname
since solaris does not have "http"")
|
|
|
7198bbb8
|
2018-10-21T03:03:57
|
|
Turn off dns/getaddrinfo_race_gotresolve by default
It is:
- pretty internal regression
- CPU bound
- right now failed on travis-ci machines
|
|
|
09c74f71
|
2018-10-21T02:50:04
|
|
Fix an error for debug locking in dns/getaddrinfo_race_gotresolve
When there is no /etc/services file evdns_getaddrinfo() will fail (with
service="ssh") and hence it will go to then "end" label with locked
rp.lock which in case of debug locking checks will bail with:
[err] ../evthread.c:220: Assertion lock->count == 0 failed in debug_lock_free
So add rp.locked flag, and unlock the lock before freeing it if it is in
locked state.
And here is how you can reproduce the issue:
$ docker run -e LD_LIBRARY_PATH=$PWD/lib -e PATH=/usr/bin:/bin:$PWD/bin -v $PWD:$PWD --rm -it debian:testing regress dns/getaddrinfo_race_gotresolve
(since debian:testing does not have /etc/services)
|
|
95918754
|
2018-05-23T15:39:13
|
|
Fix missing LIST_HEAD
Despite the presence of 'sys/queue.h' in some stdlib implementations
(i.e. uclibc) 'LIST_HEAD' macro can be missing. This fix defines this
macro in the same manner as was done previously for 'TAILQ_'.
Fixes: #539
Closes: #639 (cherry-picked)
Backport: 2.1.9
|
|
|
4c61a04c
|
2018-10-18T23:16:05
|
|
Make cmake's check_const_exists() rely on cache
|
|
|
878bb2d3
|
2018-10-17T23:21:32
|
|
Merge branch 'be-wm-overrun-v2'
* be-wm-overrun-v2:
Fix hangs due to watermarks overruns in bufferevents implementations
test: cover watermarks (with some corner cases) in ssl bufferevent
Fixes: #690
|
|
|
5a455acd
|
2018-10-17T23:21:17
|
|
Fix hangs due to watermarks overruns in bufferevents implementations
Some implementations of bufferevents (for example openssl) can overrun
read high watermark.
And after this if user callback will not drain enough data it will be
suspended (i.e. it will not be runned again anymore).
This is not the expecting behaviour as one may guess, since in this case
the data will never be read. Hence once we detected that the watermark
exceeded (even after calling user callback) we will schedule the
callback again.
This also can be fixed in bufferevent openssl implementation (by
strictly limiting how much data is added to the read buffer according to
read high watermark), but since this data is already available (and in
memory) there is no point in doing so.
|
|
|
a5b2ed56
|
2018-10-04T01:03:10
|
|
test: cover watermarks (with some corner cases) in ssl bufferevent
|
|
|
cec4a371
|
2018-10-17T11:14:48
|
|
Add -fdiagnostics-color=always for Ninja
|
|
|
8483c535
|
2018-09-14T02:24:44
|
|
evrpc: avoid NULL dereference on request is not EVHTTP_REQ_POST
Fixes: #660
|
|
7af974ee
|
2018-08-15T13:57:01
|
|
test: make sure pthread is defined
avoid warnings with any modern C99 compiler due to implicit function
declaration for pthread_create, as shown by the following :
test/regress_dns.c:2226:2: warning: implicit declaration of function
'pthread_create' is invalid in C99 [-Wimplicit-function-declaration]
THREAD_START(thread[0], race_base_run, &rp);
^
test/regress_thread.h:35:2: note: expanded from macro 'THREAD_START'
pthread_create(&(threadvar), NULL, fn, arg)
^
test/regress_dns.c:2226:2: warning: this function declaration is not a prototype
[-Wstrict-prototypes]
test/regress_thread.h:35:2: note: expanded from macro 'THREAD_START'
pthread_create(&(threadvar), NULL, fn, arg)
^
$ clang --version
Apple LLVM version 9.1.0 (clang-902.0.39.2)
Target: x86_64-apple-darwin17.7.0
Thread model: posix
InstalledDir: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin
Closes: #686 (cherry-picked)
|
|
|
29cc8386
|
2018-08-18T23:50:10
|
|
appveyor: add all build tree into artifacts in case of failure
This will help in investigations of win32 failures...
P.S. looks like there is no better way, since
"artifacts.*.publish_on_failure" hadn't been implemented yet (while was
requiested in 2015).
|
|
|
855f0804
|
2018-08-09T14:47:17
|
|
dns-example: free result in getaddrinfo callback
According to evdns.c, the result not freed by libevent after
the callback runs:
evdns_getaddrinfo_gotresolve()
{
...
data->user_cb(0, data->pending_result, data->user_data);
data->pending_result = NULL;
...
}
To reproduce, build with -fsanitize=address, add -g to the getopt
list in dns-example.c like in the current commit and run
dns-example -g google.com
Closes: #681 # cherry-picked
|
|
be371163
|
2018-06-16T17:16:10
|
|
Eliminate compiler warnings (at least for gcc/linux)
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
Closes: #646 # cherry-picked from the PR
|
|
|
ee12c516
|
2018-08-02T09:29:29
|
|
Merge branch 'evdns_getaddrinfo-race-fix'
* evdns_getaddrinfo-race-fix:
evdns: add regress test for getaddrinfo race
evdns: fix race condition in evdns_getaddrinfo()
Merges: #673
|
|
4802c132
|
2018-08-02T09:13:33
|
|
evdns: add regress test for getaddrinfo race
|
|
b59525ec
|
2018-08-02T00:35:28
|
|
evdns: fix race condition in evdns_getaddrinfo()
evdns_getaddrinfo() starts two parallel requests for A and AAAA record.
But if request is created from thread different from dns_base's, request of A record is
started immediately and may result in calling free_getaddrinfo_request() from
evdns_getaddrinfo_gotresolve() because `other_req' doesn't exist yet.
After that, request of AAAA record starts and finishes, and evdns_getaddrinfo_gotresolve()
is called again for structure that is already freed.
This commits adds locking into evdns_getaddrinfo() function.
|
|
|
e85818d2
|
2018-06-19T10:15:08
|
|
Cleanup __func__ detection
First of all __func__ is not a macro, it is char[] array, so the code
that we had before in cmake, was incorrect, i.e.:
#if defined (__func__)
#define EVENT____func__ __func__
#elif defined(__FUNCTION__)
#define EVENT____func__ __FUNCTION__
#else
#define EVENT____func__ __FILE__
#endif
So just detect do we have __func__/__FUNCTION__ in configure/cmake
before build and define EVENT__HAVE___func__/EVENT__HAVE___FUNCTION__
to use the later to choose which should be used as a __func__ (if it is
not presented).
Closes: #644
|
|
|
c3a6fe75
|
2018-08-01T09:48:42
|
|
Merge branch 'official/pr/671' -- README cleanup
* official/pr/671:
Capitalise project names consistently in README.md
Indent configure flag section to make markdown format them as code
Use https for resources that support it
Rewords awkward sentences in README.md
Fix typos in README.md
|
|
|
4728ffed
|
2018-08-01T00:58:02
|
|
autotools: include win32 specific headers for socklen_t detection on win32/mingw
The [1] removes EVENT__ prefix, and now if we will incorrectly detect
that "foobar" (or socklen_t in our case) type is not available, but
somewhere later it will be available then we will get next error [2]:
error: two or more data types in declaration specifiers
According to [3]:
- Compile something in Cygwin and you are compiling it for Cygwin.
- Compile something in MinGW and you are compiling it for Windows.
And I can confirm this, since there is _WIN32 defined (according to [4])
And since according to [5] our image in appveyour (Visual Studion 2015)
has mingw (and we use it, not cygwin) we need ws2tcpip.h (over
sys/socket.h -- which does not exist in win32) header to detect
socklen_t existence.
[1]: 587e9f5828c4eb3c6e79e9eb29f174e4ae0c05bd ("config.h can't be prefixed unconditionally")
[2]: https://ci.appveyor.com/project/nmathewson/libevent/build/job/yonukoc5q3tr3e5e#L372
[3]: https://stackoverflow.com/questions/771756/what-is-the-difference-between-cygwin-and-mingw
[4]: https://www.appveyor.com/docs/build-environment/#mingw-msys-cygwin
[5]: https://ci.appveyor.com/project/nmathewson/libevent/build/job/yonukoc5q3tr3e5e#L164
Fixes: #649
|
|
ee0c5799
|
2018-08-01T10:29:43
|
|
Capitalise project names consistently in README.md
|
|
|
72426ff4
|
2018-08-01T10:24:50
|
|
Indent configure flag section to make markdown format them as code
|
|
|
a52bd1e3
|
2018-08-01T10:21:57
|
|
Use https for resources that support it
|
|
|
560d4f32
|
2018-08-01T10:20:28
|
|
Rewords awkward sentences in README.md
|
|
|
e5816df5
|
2018-08-01T10:17:42
|
|
Fix typos in README.md
|
|
|
bdd53128
|
2018-08-01T03:01:43
|
|
appveyor: disable almost all configurations (except autotools/cmake basic)
Since we have FREE plan, we have a lot of limits on our builds, so let's
reduce their amount to make them run/start faster (I hope).
|
|
|
0f04e9a4
|
2018-08-01T00:44:58
|
|
vagrant/win32: reduce amount of threads
Otherwise I got:
win: 855 [main] sh 840 fork: child -1 - CreateProcessW failed for 'C:\tools\cygwin\bin\sh.exe', errno 5
win: /bin/sh: /usr/bin/sed: Input/output error
win: /bin/sh: line 1: /bin/sh: Input/output error
win: CC bufferevent_pair.lo
win: /bin/sh: fork: Input/output error
win: make[1]: *** [Makefile:1722: buffer.lo] Error 126
win: make[1]: *** Waiting for unfinished jobs....
win: CC bufferevent_ratelim.lo
win: 176 [main] sh 2440 fork: child -1 - CreateProcessW failed for 'C:\tools\cygwin\bin\sh.exe', errno 5
win: /bin/sh: fork: Input/output error
win: 0 [main] sh 3684 fork: child -1 - CreateProcessW failed for 'C:\tools\cygwin\bin\sh.exe', errno 5
win: ./libtool: fork: Input/output error
win: make[1]: *** [Makefile:1722: bufferevent_filter.lo] Error 254
win: 0 [main] sh 1356 fork: child -1 - CreateProcessW failed for 'C:\tools\cygwin\bin\sh.exe', errno 5
win: ./libtool: fork: Input/output error
win: 0 [main] sh 3640 fork: child -1 - CreateProcessW failed for 'C:\tools\cygwin\bin\sh.exe', errno 5
win: ./libtool: line 922: /usr/bin/sed: Input/output error
win: ./libtool: fork: Input/output error
win: 0 [main] sh 3444 fork: child -1 - CreateProcessW failed for 'C:\tools\cygwin\bin\sh.exe', errno 5
win: 0 [main] sh 3796 fork: child -1 - CreateProcessW failed for 'C:\tools\cygwin\bin\sh.exe', errno 5
win: ./libtool: fork: Input/output error
win: 693 [main] sh 304 fork: child -1 - CreateProcessW failed for 'C:\tools\cygwin\bin\sh.exe', errno 5
win: ./libtool: fork: Input/output error
win: ./libtool: fork: Input/output error
win: ./libtool: line 1377: /usr/bin/sed: Input/output error
win: : error: Failed to create '.libs'
|
|
|
9cd279c7
|
2018-08-01T00:28:48
|
|
vagrant/win32: add missing aclocal
|
|
30d77f1b
|
2018-07-05T11:25:24
|
|
Fix typo
Closes: #658
|
|
514dc757
|
2018-07-05T12:46:51
|
|
[http] fix C90 warnings
|
|
|
24236aed
|
2018-06-20T23:47:51
|
|
Tell Travis to use quiet git clone
Otherwise the logs show a bunch of pointless progress, which is
really only useful for interactive sessions.
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
Closes: #651 (cherry-picked)
|
|
d9ffd221
|
2018-05-15T12:14:05
|
|
test: make regress_dns C89 compatible
Closes: #635 (cherry-picked)
|
|
|
064a5987
|
2018-06-21T02:37:24
|
|
Ignore evconfig-private.h for autotools
Fixes: e2874d95af584b3657eb5dc1d2e5404811dce117 ("Adopt ignore rules for
cmake + ninja")
|
|
|
35258a1f
|
2018-06-19T10:18:46
|
|
Replace all EVENT_HAVE with EVENT__HAVE (EVENT__HAVE_USLEEP)
|
|
|
d2acf67e
|
2018-06-16T17:39:34
|
|
Add convenience macros for user-triggered events
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
Closes: #647 (picked)
|
|
|
755896ef
|
2018-06-16T16:44:06
|
|
Get rid of macros which are never used
There are no expansions of these macros or tests for their existence.
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
|
|
0789bc52
|
2018-05-25T18:07:28
|
|
fix spelling mistakes
Closes: nmathewson/Libevent#162
|
|
|
587e9f58
|
2018-04-18T15:09:35
|
|
config.h can't be prefixed unconditionally
The script make-event-config.sed was mangling all the symbols by
prefixing them with "EVENT__". The problem here is that some
symbols aren't for local consumption within libevent, but rather
influence other system header files (ex: __USE_FILE_OFFSET64 is
used by dozens of header files including <sys/sendfile.h>).
As a workaround, all symbols starting with a capital letter only
(with the exception of STDC_HEADERS which must also be left
untouched) will be mangled.
Future contributors will need to be aware of this distinction.
|
|
|
c57f5c34
|
2018-05-07T02:39:44
|
|
Make rpc headers self-compilable
Fixes: #633
|
|
|
1619cc35
|
2018-05-02T19:34:04
|
|
Merge pull request #632 from NathanFrench/realign_struct_event
Reorder fields in struct event for memory efficiency
|
|
|
1af80176
|
2018-04-30T18:13:45
|
|
[core] re-order fields in struct event for memory efficiency
The sizeof `struct event` can reduced on both 32 bit and 64 bit systems
by moving the 4 bytes that make up `ev_events` and `ev_res` below `ev_fd`,
before `struct event_base * ev_base;` since our compiler wouldn't dare do
such a thing (it instead will pad twice, whereas it only needs to be padded
once)
```C
struct event {
/* OFFS | SZ Bytes | Total Bytes | START - END */
struct event_callback ev_evcallback; /* 0x0 | 40 | 40 | 0x0 - 0x28 */
union { /* 0x28 | ----------- | ----------- | ------------ */
TAILQ_ENTRY(event) ev_next_with_common_timeout; /* | ((16)) | | */
int min_heap_idx; /* | ((04)) | | */
} ev_timeout_pos; /* | 16 | 56 | 0x28 - 0x38 */
int ev_fd; /* 0x38 | 04 | 60 | 0x38 - 0x3c */
```
Since the next field is 8 bytes in length, and we are up to 60 bytes, `ev_fd` ends up being padded (4 more bytes on 64b).
```C
/* --- 1 byte gap HERE ---> 1 | <61> */
/* --- 1 byte gap HERE ---> 1 | <62> */
/* --- 1 byte gap HERE ---> 1 | <63> */
/* --- 1 byte gap HERE ---> 1 | <64> */
struct event_base * ev_base; /* 0x3c | 8 | 68 | 0x3c - 0x40 */
union { /* 0x40 | ------------ | ---------- | ------------ */
struct { /* | ------------ | | */
LIST_ENTRY (event) ev_io_next; /* | ((16+ | | */
struct timeval ev_timeout; /* | 16)) | | */
} ev_io; /* | ((32)) | | */
struct { /* | ------------ | | */
LIST_ENTRY (event) ev_signal_next; /* | ((16+ | | */
short ev_ncalls; /* | 02+ | | */
short * ev_pncalls; /* | 08)) | | */
} ev_signal; /* | ((26)) | | */
} ev_; /* 0x60 | 32 | 100 | 0x40 - 0x60 */
short ev_events; /* 0x60 | 2 | 102 | 0x60 - 0x62 */
short ev_res; /* 0x62 | 2 | 104 | 0x62 - 0x64 */
```
We now hit another line, `struct timeval` is 16 bytes on 64b arch, so we have 4 more bytes
of padding on `ev_res`.
```C
/* --- 1 byte gap HERE --- */
/* --- 1 byte gap HERE --- */
/* --- 1 byte gap HERE --- */
/* --- 1 byte gap HERE --- */
struct timeval ev_timeout; /* 0x64 | 16 | 120 | 0x64 - 0x74 */
};
```
After moving `ev_events` and `ev_res` below `ev_fd` we have something
a bit more optimal:
```C
struct event2 {
/* OFFS | SZ / Bytes | RSUM Bytes | START - END */
struct event_callback ev_evcallback; /* 0x0 | 40 | 40 | 0x0 - 0x28 */
union { /* 0x28 | ------------ | ---------- | ------------ */
TAILQ_ENTRY(event) ev_next_with_common_timeout; /* | ((16)) | | */
int min_heap_idx; /* | ((04)) | | */
} ev_timeout_pos; /* | 16 | 56 | 0x28 - 0x38 */
int ev_fd; /* 0x38 | 4 | 60 | 0x38 - 0x3c */
short ev_events; /* 0x3c | 2 | 62 | 0x3c - 0x3e */
short ev_res; /* 0x3e | 2 | 64 | 0x3e - 0x40 */
struct event_base * ev_base; /* 0x40 | 8 | 74 | 0x40 - 0x48 */
union { /* 0x48 | ------------ | ---------- | ------------ */
struct { /* | ------------ | | */
LIST_ENTRY (event) ev_io_next; /* | ((16+ | | */
struct timeval ev_timeout; /* | 16)) | | */
} ev_io; /* | ((32)) | | */
struct { /* | ------------ | | */
LIST_ENTRY (event) ev_signal_next; /* | ((16+ | | */
short ev_ncalls; /* | 02+ | | */
short * ev_pncalls; /* | 08)) | | */
} ev_signal; /* | ((26)) | | */
} ev_; /* | 32 | 106 | 0x48 - 0x68 */
struct timeval ev_timeout; /* 0x68 | 16 | 120 | 0x68 - 0x78 */
};
```
We still have a gap here, but the first was removed.
Again, we can save 8 bytes on both 32 and 64 word sizes (32/64 byte cacheline).
Below are the results for testing v2.1.6 -> master -> master + this patch (Release/-O3)
Code:
```C
#include <event2/event.h>
int
main(int argc, char ** argv) {
printf("%zu\n", event_get_struct_event_size());
return 0;
}
```
Branch: `master` (2.2.x)
```
$ gcc -O3 -Wall -Wl,-R/usr/local/lib bleh.c -L/usr/local/lib -o bleh -levent
$ ldd bleh
linux-vdso.so.1 => (0x00007ffc3df50000)
libevent.so.2.2.0 => /usr/local/lib/libevent.so.2.2.0 (0x00007f91fd781000)
libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007f91fd3a1000)
libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x00007f91fd182000)
/lib64/ld-linux-x86-64.so.2 (0x00007f91fdbcc000)
$ ./bleh
128
```
Release: `2.1.6`
```
$ gcc -O3 bleh.c -o bleh -levent
$ ldd bleh
linux-vdso.so.1 => (0x00007ffd43773000)
libevent-2.1.so.6 => /usr/lib/x86_64-linux-gnu/libevent-2.1.so.6 (0x00007feb3add6000)
libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007feb3a9f6000)
libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x00007feb3a7d7000)
/lib64/ld-linux-x86-64.so.2 (0x00007feb3b22a000)
$ ./bleh
128
```
Branch: `this one`
```
$ gcc -O3 -Wl,-R./lib bleh.c -o bleh -L./lib -levent
$ ldd bleh
linux-vdso.so.1 => (0x00007ffff55f7000)
libevent.so.2.2.0 => ./lib/libevent.so.2.2.0 (0x00007ff8e5c82000)
libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007ff8e58a2000)
libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x00007ff8e5683000)
/lib64/ld-linux-x86-64.so.2 (0x00007ff8e60cd000)
$ ./bleh
120
```
|
|
|
23c2914f
|
2018-04-24T00:59:11
|
|
Notify event base if there are no more events, so it can exit without delay
Fixes: #623
|
|
|
accf383e
|
2018-04-23T01:50:55
|
|
Fix CheckFunctionExistsEx() cmake macro on win32
For example under mingw64 it could not detect that strtok_r() exists,
because it checks with:
void *p = func_name;
And for this you need the function to be defined, so just sync our
CheckFunctionExistsEx.c with CheckFunctionExists.c from cmake (and later
we should drop them out) since it does correct things to detech
functions existence.
Also for WIN32 there is -FIwinsock2.h -FIws2tcpip.h, and I guess that is
not works for mingw gcc (since -F in gcc is framework, and in windows
-FI is like -include in gcc). But looks like we do not need them
already (due to fixed CheckFunctionExistsEx()).
Refs: #605
|
|
|
d1c8993c
|
2018-04-24T14:46:06
|
|
test/dns: install correct RLIMIT_NOFILE in bufferevent_connect_hostname_emfile
Otherwise poll() will fail with EINVAL:
EINVAL The nfds value exceeds the RLIMIT_NOFILE value.
P.S. and cleanup this test a little, with early-return.
CI: https://travis-ci.org/libevent/libevent/jobs/370350426
|
|
|
da028dec
|
2018-04-24T02:56:08
|
|
travis-ci: allow tests under osx to fail (temporary, until we will fix them)
|
|
|
e4edc7fc
|
2018-04-24T02:22:58
|
|
test/http: cover evhttp_connection_get_addr() for incomming connections
Refs: #510
|
|
367cd9e5
|
2018-02-11T16:28:58
|
|
Fix evhttp_connection_get_addr() fox incomming http connections
Install conn_address of the bufferevent on incomping http connections
(even though this is kind of subsytem violation, so let's fix it in a
simplest way and thinkg about long-term solution).
Fixes: #510
Closes: #595 (pick)
|
|
|
623ef3cc
|
2018-04-24T01:34:06
|
|
test/dns: cleanup test_bufferevent_connect_hostname()
|
|
|
a3d8f2e0
|
2018-04-24T01:45:00
|
|
test/dns: verify bufferevent_socket_connect() errorcb invoking if socket() fails
Refs: #600
|
|
f7bc1337
|
2018-02-23T19:15:12
|
|
bufferevent_socket_connect{,_hostname}() missing event callback and use ret code
- When socket() failed in bufferevent_socket_connect() , the event
callback should be called also in
bufferevent_socket_connect_hostname(). eg. when use
bufferevent_socket_connect_hostname() to resolve and connect an IP
address but process have a smaller ulimit open files, socket() fails
always but caller is not notified.
- Make bufferevent_socket_connect()'s behavior more consistent: function
return error then no callback, function return ok then error passed by
event callback.
Fixes: #597
Closes: #599
Closes: #600
|
|
|
a5f19422
|
2018-04-23T00:52:57
|
|
Merge branch 'pull-628'
* pull-628:
Give priority to the build directory headers
Do not ship evconfig-private.h in dist archive
Closes: #628
|
|
5d74565a
|
2018-04-19T09:39:39
|
|
Give priority to the build directory headers
Not changing anything right now AFAIK. But if for any reason in the
future we end up with two headers with the same name in the source and
build directories, chances are we want to use the one in the build
directory.
|
|
|
0379eb4b
|
2018-04-19T09:28:36
|
|
Do not ship evconfig-private.h in dist archive
It will be generated by autotools, so there is not reason to include it.
And infact this breaks compilation with out-of-tree builds (VPATH),
since, for the quote form of the include directive, headers in the
directory of the file with the #include line have priority over those
named in -I options, the copy of evconfig-private.h from the source
directory had priority over the one in the build directory.
|
|
|
e2874d95
|
2018-04-23T00:26:08
|
|
Adopt ignore rules for cmake + ninja
In case we have build directory differs from source directory there will be
bunch of files we should ignore, so just remove leading "/" for some or rules.
And fix others.
|
|
ab3224c3
|
2018-04-20T15:48:32
|
|
Fix assert() condition in evbuffer_drain() for IOCP
In the case of iocp, in the for loop above, there is a situation where:
remaining == chain->off == 0
And this happens due to CHAIN_PINNED_R() case (that is used only in
buffer_iocp.c)
Closes: #630 (picked)
|
|
ba78ba9e
|
2018-04-15T23:02:05
|
|
Add Gitter badge
|
|
0fa43c99
|
2018-04-05T15:17:06
|
|
cmake: ensure windows dll's are installed as well as lib files
Closes: #621
|
|
791e3de0
|
2018-04-03T15:43:22
|
|
Generating evdns_base_config_windows_nameservers docs on all platforms
|
|
|
2c156294
|
2018-04-03T15:17:51
|
|
Fixing doxygen docs for evdns_base_search_clear when generated on non-windows machines
|
|
28b80754
|
2018-04-02T13:18:27
|
|
Fix build with LibreSSL 2.7
LibreSSL 2.7 implements OpenSSL 1.1 API except for BIO_get_init()
See also: https://bugs.freebsd.org/226900
Signed-off-by: Bernard Spil <brnrd@FreeBSD.org>
Closes: #617 (cherry-pick)
|
|
|
33baa4e5
|
2018-03-28T08:25:28
|
|
Avoid possible SEGVs in select() (in unit tests)
Per the POSIX definition of select():
http://pubs.opengroup.org/onlinepubs/009696699/functions/pselect.html
"Upon successful completion, the select() function may modify the object
pointed to by the timout argument."
If "struct timeval" pointer is a "static const", it could potentially
be allocated in a RO text segment. The kernel would then try to copy
back the modified value (with the time remaining) into a read-only
address and SEGV.
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
Closes: #614
|
|
|
4ba48739
|
2018-03-20T18:04:53
|
|
don't fail be_null_filter if bytes are copied
Otherwise it will not reset timeouts for the IO, but other things should
still works correctly.
Also evbuffer_remove_buffer() could return 0, due to empty buffers, for
example during flushing bufferevent, so let's count this is BEV_OK too.
|
|
|
40550814
|
2018-03-20T15:12:07
|
|
Call underlying bev ctrl GET_FD on filtered bufferevents
Fixes: #611
Fixes: #610
|
|
|
08a0d366
|
2018-02-27T21:12:14
|
|
Fix base unlocking in event_del() if event_base_set() runned in another thread
Image next situation:
T1: T2:
event_del_()
lock the event.ev_base.th_base_lock
event_del_nolock_() event_set_base()
unlock the event.ev_base.th_base_lock
In this case we will unlock the wrong base after event_del_nolock_()
returns, and deadlock is likely to happens, since event_base_set() do
not check any mutexes (due to it is possible to do this only if event is
not inserted anywhere).
So event_del_() has to cache the base before removing the event, and
cached base.th_base_lock after.
|
|
|
f0fd92f2
|
2018-02-27T21:08:35
|
|
Convert event_debug_*() helpers from macros to static functions
|
|
1a448088
|
2017-05-14T06:22:11
|
|
Provide Makefile variables LIBEVENT_{CFLAGS,CPPFLAGS,LDFLAGS}
This will allow a parent package to specify compiler, CPP, and linker
flags to a libevent built as a sub-package. Document this in
whatsnew-2.2.txt.
Picked-from: #506
Closes: #506
|
|
|
cd57e38c
|
2018-02-13T17:43:02
|
|
http: remove message in case !Content-Length and Connection!=close
Since [1] GET can have body, and hence for every incomming connection it
will print this error.
[1] db483e3b002b33890fc88cadd77f6fd1fccad2d2 ("Allow bodies for
GET/DELETE/OPTIONS/CONNECT")
Noticed-by: BotoX (irc)
Refs: #408
|
|
|
93913da1
|
2018-02-12T19:43:13
|
|
buffer: fix incorrect unlock of the buffer mutex (for deferred callbacks)
TSAN reports:
WARNING: ThreadSanitizer: unlock of an unlocked mutex (or by a wrong thread) (pid=17111)
#0 pthread_mutex_unlock /build/gcc/src/gcc/libsanitizer/sanitizer_common/sanitizer_common_interceptors.inc:3621 (libtsan.so.0+0x00000003b71c)
#1 evbuffer_add <null> (libevent_core-2.2.so.1+0x00000000ddb6)
...
Mutex M392 (0x7b0c00000f00) created at:
#0 pthread_mutex_init /build/gcc/src/gcc/libsanitizer/tsan/tsan_interceptors.cc:1117 (libtsan.so.0+0x0000000291af)
#1 <null> <null> (libevent_pthreads-2.2.so.1+0x000000000d46)
...
$ addr2line -e /lib/libevent_core-2.2.so.1 0x00000000ddb6
/src/libevent/buffer.c:1815 (discriminator 1)
Introduced-in: ae2b84b2575be93d0aebba5c0b78453836f89f3c ("Replace
deferred_cbs with event_callback-based implementation.")
|
|
f24b28e4
|
2018-01-15T17:30:08
|
|
Fix typos in comments
|
|
|
61c21492
|
2018-01-09T21:44:57
|
|
http: fix leaks in evhttp_uriencode()
Fixes: #584
|
|
|
416b48ba
|
2018-01-04T19:28:59
|
|
Merge branch 'listener-immediate-close'
* listener-immediate-close:
test/listener: cover immediate-close logic
Immediately stop trying to accept more connections if listener disabled
|
|
|
cb6995cf
|
2018-01-04T19:26:50
|
|
test/listener: cover immediate-close logic
|
|
e03cd0b7
|
2017-12-17T22:43:00
|
|
Immediately stop trying to accept more connections if listener disabled
This is a refined version of the logic previously in #578
The rationale is that the consumer of sockets may wish to temporarily
delay accepting for some reason (e.g. being out of file-descriptors).
The kernel will then queue them up. The kernel queue is bounded and
programs like NodeJS will actually try to quickly accept and then close
(as the current behaviour before this PR).
However, it seems that libevent should allow the user to choose whether
to accept and respond correctly if the listener is disabled.
|
|
88a1abe8
|
2017-12-18T14:01:15
|
|
Fixed last_with_datap description
Grammar and typo fix
|
|
|
727bcea1
|
2017-12-01T01:29:32
|
|
http: add callback to allow server to decline (and thereby close) incoming connections.
This is important, as otherwise clients can easily exhaust the file
descriptors available on a libevent HTTP server, which can cause
problems in other code which does not handle EMFILE well: for example,
see https://github.com/bitcoin/bitcoin/issues/11368
Closes: #578 (patch cherry picked)
|
|
|
6e5c15d0
|
2017-12-16T23:28:41
|
|
Fix typo in cmake because of which EVENT__SIZEOF_SIZE_T was wrong
Fixes: #580
|
|
|
65eb529a
|
2017-12-02T12:53:57
|
|
CONNECT method only takes an authority
|
|
6cce7458
|
2017-12-09T14:39:07
|
|
If precise_time is false, we should not set EVENT_BASE_FLAG_PRECISE_TIMER
Fixes: 630f077c296de61c7b99ed83bf30de11e75e2740 ("Simple unit tests for
monotonic timers")
|
|
|
bc65ffc1
|
2017-12-10T23:57:19
|
|
Merge branch 'evconnlistener-do-not-close-client-fd'
Fixes: #577
* evconnlistener-do-not-close-client-fd:
listener: cover closing of fd in case evconnlistener_free() called from acceptcb
Revert "Fix potential fd leak in listener_read_cb()"
|
|
|
8a460e38
|
2017-12-05T23:26:28
|
|
listener: cover closing of fd in case evconnlistener_free() called from acceptcb
|
|
|
94eae336
|
2017-12-05T22:59:20
|
|
Revert "Fix potential fd leak in listener_read_cb()"
This reverts commit a695a720cda892c270736d127333d73553842094.
@kgraefe:
"I believe that this commit is just wrong: if lev->cnt is not 1 after
the callback, new_fd will still never be closed in listener_read_cb().
So in that case it is the responsibility of the user's code to close
the file descriptor (which is fine). But why shouldn't it be in the
other case? And how does the user's code know?"
And I agree
|
|
6ee73ea9
|
2017-11-29T11:13:51
|
|
Fix generation of LibeventConfig.cmake for the installation tree
'LIBEVENT_INCLUDE_DIRS' is properly initialized in 'LibeventConfig.cmake' as
'LibeventConfig.cmake.in' contains usage of 'LIBEVENT_CMAKE_DIR' and
'EVENT_INSTALL_INCLUDE_DIR' variables but not 'EVENT_CMAKE_DIR' and
'EVENT__INCLUDE_DIRS'.
Related typos are fixed.
|
|
|
c2c08e02
|
2017-11-22T10:33:15
|
|
Add missing includes into openssl-compat.h
Before it depends from the caller #include appropriate headers (at least
for OPENSSL_VERSION_NUMBER), but let's make it independent.
Fixes: #574
|
|
|
33e363f3
|
2017-11-20T02:07:54
|
|
Free dns/event bases in dns-example to avoid leaks
|
|
b49c70cc
|
2017-11-05T12:18:49
|
|
Fix incorrect ref to evhttp_get_decoded_uri in http.h
Replaces reference in the http.h include header file to evhttp_get_decoded_uri
with evhttp_uridecode. There is no function called evhttp_get_decoded_uri.
|
|
|
306747e5
|
2017-11-04T19:13:28
|
|
Fix crashing http server when callback do not reply in place from *gencb*
This is the second hunk of the first patch
5ff8eb26371c4dc56f384b2de35bea2d87814779 ("Fix crashing http server when
callback do not reply in place")
Fixes: #567
|