kmx git

Commit	Date	Message
80852425	2017-06-30T02:27:08	Document some obvious cases where a function might also return NULL Closes: #525
cd285e42	2017-05-29T22:11:48	Fix event_debug_logging_mask_ exporting on win32
1771ec0f	2017-05-29T22:12:09	Vagrant test
3ac4f176	2017-05-29T22:07:57	Revert "Make event_debug_get_logging_mask_() real symbol (win32)" This reverts commit 5f17c12de021c070501e97268e595da82bf38879.
5f17c12d	2017-05-29T20:46:33	Make event_debug_get_logging_mask_() real symbol (win32) I cannot made it work without this, even though dumpbin shows that that symbol exists in the event_core.dll, event_extra.dll failed to compile: ==> win: Creating library C:/vagrant/.cmake-vagrant/lib/Debug/event_extra.lib and object C:/vagrant/.cmake-vagrant/lib/Debug/event_extra.exp ==> win: http.obj : error LNK2001: unresolved external symbol _event_debug_logging_mask_ [C:\vagrant\.cmake-vagrant\event_extra_shared.vcxproj] ==> win: C:\vagrant\.cmake-vagrant\bin\Debug\event_extra.dll : fatal error LNK1120: 1 unresolved externals [C:\vagrant\.cmake-vagrant\event_extra_shared.vcxproj] ==> win: Done Building Project "C:\vagrant\.cmake-vagrant\event_extra_shared.vcxproj" (default targets) -- FAILED. And dumpbin: sh-4.1$ /cygdrive/c/Program\ Files\ \(x86\)/Microsoft\ Visual\ Studio\ 12.0/VC/bin/amd64/dumpbin.exe /EXPORTS ./bin/Debug/event_core.dll M 202 C9 00059A3C event_debug_logging_mask_ = _event_debug_logging_mask_
ce3af533	2017-05-29T15:04:50	Fix visibility issues under (mostly on win32) Refs: #511 Fixes: 7182c2f561570cd9ceb704623ebe9ae3608c7b43 ("cmake: build SHARED and STATIC libraries (like autoconf does)")
8b29b136	2017-05-29T16:36:24	cmake: fix building bench/bench_cascade under win32 (missing getopt) Because of typo in cmake, now rewrote to make it less error prone (since even for non-win32 there are variables overlap).
346f8e17	2017-05-29T16:32:30	cmake: avoid requiring cmake 3.1 on win32 for dns-example Avoid using target_sources() from cmake 3.1, just pass needable sources as list.
2773a5ed	2017-05-29T18:48:48	cmake: set CMP0054 to NEW to avoid variables over expansion (since cmake 3.8) ==> win: CMake Warning (dev) at CMakeLists.txt:782 (elseif): ==> win: Policy CMP0054 is not set: Only interpret if() arguments as variables or ==> win: keywords when unquoted. Run "cmake --help-policy CMP0054" for policy ==> win: details. Use the cmake_policy command to set the policy and suppress this ==> win: warning. ==> win: ==> win: Quoted variables like "MSVC" will no longer be dereferenced when the policy ==> win: is set to NEW. Since the policy is not set the OLD behavior will be used. ==> win: This warning is for project developers. Use -Wno-dev to suppress it.
74a2f593	2017-05-29T16:24:48	le-proxy: fix compiling under win32 (because of no "inline" in C) Include <util-internal.h> for regular "inline", since it is used everywhere.
d22e12ee	2017-05-28T16:38:20	http-connect: exit on error
15da23cf	2017-05-01T21:56:27	test: do not use .fieldname in structure initializations (fixes win32) Fixes: #497
8d89c212	2017-05-01T21:55:00	Add missing print-winsock-errors.c into dist archive Refs: #497
00ae45fe	2017-04-25T23:10:13	le-proxy: add missing return statement (UAF) Refs: #501501501501501501501501501
0b4b0efd	2015-04-13T08:54:52	Return from event_del() after the last event callback termination Delete the event from the queue before blocking for the current event callback termination. Ensures that no callback is being executed when event_del() returns, hence making this function a secure mechanism to access data which is handled in the event callack. Fixes: #236 Fixes: #225 Refs: 6b4b77a Fixes: del_wait
5ff83989	2017-04-30T01:11:31	test: cover that after event_del() callback will not be scheduled again Refs: #236 Refs: #225
f988c5ca	2017-04-28T19:36:42	Correct URL for the libevent logo
c6c74ce2	2017-04-20T18:03:50	Explicitly call SSL_clear when reseting the fd. If reconnecting the via BEV_CTRL_SET_FD, bufferevent_openssl.c expects OpenSSL to reuse the configuration state in the SSL object but retain connection state. This corresponds to the SSL_clear API. The code currently only calls SSL_set_connect_state or SSL_set_accept_state. Due to a quirk in OpenSSL, doing this causes the handshake to implicitly SSL_clear the next time it is entered. However, this, in the intervening time, leaves the SSL object in an odd state as the connection state has not been dropped yet. This behavior also does not appear to be documented by OpenSSL. Instead, call SSL_clear explicitly: https://www.openssl.org/docs/manmaster/man3/SSL_clear.html
92cc0b9c	2017-04-13T14:58:13	bufferevent: refactor to use type check macros
266f43af	2017-03-27T15:50:23	Fix arc4random_addrandom() detecting and fallback (regression) But this is kind of hot-fix, we definitelly need more sane arc4random compat layer. Fixes: #488 Introduced-in: 6541168 ("Detect arch4random_addrandom() existence")
177e2171	2017-03-24T16:49:59	Make event_count macros cleaner Fixes: #489
2e52bace	2017-03-16T23:32:43	fix evutil_make_internal_pipe_'s comment that fd[0] for read and f[1] for write
b1e8a413	2017-03-14T13:33:31	cmake: use APPEND during exporting targets (for old cmake) On centos with cmake 2.8.12.2: CMake Error at cmake/AddEventLibrary.cmake:92 (export): export called with target "event_extra_shared" which requires target "event_core_shared" that is not in the export list. If the required target is not easy to reference in this call, consider using the APPEND option with multiple separate calls. But on newer cmake I guess everything is ok. Fixes: 7182c2f561570cd9ceb704623ebe9ae3608c7b43 ("cmake: build SHARED and STATIC libraries (like autoconf does)")
d84f0205	2017-03-14T13:31:02	Merge branch 'win32-fixes' This patchset fixes win32 builds after some previous patches (referenced in particular commits), and also removes some quirks for win32. * win32-fixes: test: do not return void log/win32: fix exporting extern variable log-internal: missing extern "C" log: remove USE_GLOBAL_FOR_DEBUG_LOGGING cmake: Export missing symbols for win32 cmake: eliminate EVENT_BUILDING_REGRESS_TEST, since we link with shared libs test: windows doesn't have WNOWAIT cmake: clean not used #defines from event-config.h cmake: add <pthread.h> only for non-win32
8acfb0cd	2017-03-14T13:21:16	test: do not return void
92cf234b	2017-03-14T00:33:26	log/win32: fix exporting extern variable ==> win: C:\vagrant\log.c(73): error C2370: 'event_debug_logging_mask_' : redefinition; different storage class [C:\vagrant\.cmake-vagrant\event_core_shared.vcxproj]
a9c2116f	2017-03-14T01:06:10	log-internal: missing extern "C"
49dcf4d5	2017-03-13T22:33:38	log: remove USE_GLOBAL_FOR_DEBUG_LOGGING Fixes: 9081b66c9c1446b11c0052bade799fafb460efed ("Export symbols for -fvisibility=hidden (under cmake)")
48ead58e	2017-03-14T00:46:47	cmake: Export missing symbols for win32
30f2a969	2017-03-14T00:07:17	cmake: eliminate EVENT_BUILDING_REGRESS_TEST, since we link with shared libs Before 7182c2f561570cd9ceb704623ebe9ae3608c7b43 ("cmake: build SHARED and STATIC libraries (like autoconf does)") it links with *.c.
a4b1828d	2017-03-13T23:52:15	test: windows doesn't have WNOWAIT Fixes: 66a4eb0c3ae3b1f22b084b2d3aeb5c872f37efbd ("Check for WNOWAIT in waitpid() in runtime (not in cmake/configure)")
d5e10870	2017-03-13T21:54:33	cmake: clean not used #defines from event-config.h - _LARGE_FILES: we have it in evconfig-private.h.in, drop from event-config.h - EVENT___FILE_OFFSET_BITS: remove it for now - EVENT__const
2d2299cf	2017-03-13T22:02:45	cmake: add <pthread.h> only for non-win32 Fixes: b2b4b4d74e78a3e4fe7a74224d1e6aada5bde351 ("cmake: add <pthread.h> into CMAKE_REQUIRED_INCLUDES for sizeof(pthread_t)")
882f537c	2017-03-13T21:40:52	cmake: fix pkgconfig generation (copy-paste typo)
489991a2	2017-03-13T21:05:32	Merge branch 'cmake-missing-bits' This patchset adds next missing things (in compare to autotools): - pkgconfig - event_pthreads/event_openssl - compile shared/static libraries always And some fixes, because it will not build after fixing other things: - export missing symbols for cmake (-fvisibility=hidden) * cmake-missing-bits: cmake: support visibility for AppleClang too cmake: fix export absolute path and relative path and cleanup a bit cmake: generate and install pkgconfig files cmake: build SHARED and STATIC libraries (like autoconf does) cmake: add missing event_openssl/event_pthreads libraries Export symbols for -fvisibility=hidden (under cmake) Refs: #246
805e1aa8	2017-03-13T15:49:11	cmake: support visibility for AppleClang too Defaults apple linker behaviour is -two_levelnamespace, and you cannot use "-undefined suppress" with it, so let's link non event_core with event_core library to avoid undefined symbols.
fd5a3fc0	2017-03-13T03:18:02	cmake: fix export absolute path and relative path and cleanup a bit Remove next vars, since I don't think that somebody want to change it: - bin - lib - include And fix exports problem: CMake Error: INSTALL(EXPORT "LibeventTargets") given absolute DESTINATION "/usr/lib/cmake/libevent" but the export references an installation of target "event_core_static" which has relative DESTINATION "lib".
86c73575	2017-03-13T02:26:39	cmake: generate and install pkgconfig files Fixes: #443
7182c2f5	2017-03-12T23:31:59	cmake: build SHARED and STATIC libraries (like autoconf does) Since they are useful for debugging, and if autotools build them then cmamke has to do this too, to make migration more simple. And now: - tests: uses shared libraries (since this is upstreams one) - other binaries: uses static libraries This removes next private config: - EVENT__NEED_DLLIMPORT
72ef9d16	2016-11-07T00:46:45	cmake: add missing event_openssl/event_pthreads libraries This will remove openssl requirement if you don't use it (i.e. if you not link with openssl_pthreads). Plus it fixes some linking dependencies: - libm required only for test-ratelim And fix some coding style alignment issues. Refs: #246
9081b66c	2017-03-12T20:50:35	Export symbols for -fvisibility=hidden (under cmake) Fixes: #442
83b15847	2017-03-12T13:12:53	travis-ci: do not run coverage on osx First of all we don't really need two of them, and also with apple-clang it will fail (because of our checks in cmake), so just exclude it from the build matrix.
7004ee8e	2017-03-09T09:32:43	travis-ci: use travis_retry to reduce number of problems with flacky tests
e01b9937	2017-03-08T23:48:30	appveyor: remove -DEVENT__ENABLE_VERBOSE_DEBUG=ON build See-also: c199df7bc78824ff579ff34c5f9f922034e8fa31 ("travis-ci: remove -DEVENT__ENABLE_VERBOSE_DEBUG=ON (too much useless logs)")
65870949	2017-03-12T01:00:41	Fix detection of timerfd_create() in CMake.
819d0493	2017-03-08T23:15:33	Do not add epoll_sub (syscall wrappers) for epoll in cmake Nowadays mostly all supported OS'es has this wrappers, and some of them (like SmartOS) has wrappers but doesn't have __NR_epoll* defines for syscall numbers, so just drop them (instead of adding yet another check int cmake like autotools has, since this will break building in cross-compile environment). Also one minor note, configure doesn't added epoll_sub.c either, since it check epoll_create() in runtime. And I tested it in SmartOS and it even works (`make verify`). Fixes: #463
5aade2d3	2017-03-08T17:58:26	Merge branch 'cmake-configure-fixes-v2' Fixes in cmake, to make it more like configure and support some cross-compiling. * cmake-configure-fixes-v2: cmake: fix extracting of the version from git (check for number of matches) Detect arch4random_addrandom() existence Use off_t instead of ev_off_t for sendfile() (fixes android build) cmake: detect _GNU_SOURCE not by __GNU_LIBRARY__ only (fallback to _GNU_SOURCE) Check for WNOWAIT in waitpid() in runtime (not in cmake/configure) cmake: add <pthread.h> into CMAKE_REQUIRED_INCLUDES for sizeof(pthread_t) cmake: fix values for #cmakedefine cmake: drop duplicates from event-config template cmake: add value for the #cmakedefine macros (like autoconf) cmake: Fix checking of enum values from sysctl.h
d65e3e1d	2017-03-06T14:48:56	cmake: fix extracting of the version from git (check for number of matches)
6541168d	2017-03-06T00:55:16	Detect arch4random_addrandom() existence Refs: #370 Refs: #475
f479a008	2017-03-06T00:45:09	Use off_t instead of ev_off_t for sendfile() (fixes android build) Refs: #475
71a56bd3	2017-03-06T00:30:21	cmake: detect _GNU_SOURCE not by __GNU_LIBRARY__ only (fallback to _GNU_SOURCE) Fixes pipe2() on Android-9/armv7-a Refs: #475
66a4eb0c	2017-03-06T00:05:50	Check for WNOWAIT in waitpid() in runtime (not in cmake/configure) Because checking in cmake breaks cross-compiling. Introduced-in: 43eb56c7c738e3642f0981e3dd6ab9e082eec798. Fixes: #482 Fixes: #462 Refs: #475 v2: use waitid() with WNOWAIT v3: use WNOWAIT only if it available in waitpid(), because not all netbsd supports it
b2b4b4d7	2017-03-05T22:07:23	cmake: add <pthread.h> into CMAKE_REQUIRED_INCLUDES for sizeof(pthread_t) Fixes: #475
7f945b2d	2017-03-05T21:58:20	cmake: fix values for #cmakedefine
85272602	2017-03-05T21:57:45	cmake: drop duplicates from event-config template
336ab816	2017-03-05T21:48:56	cmake: add value for the #cmakedefine macros (like autoconf) Before this patch for cmake we had: #define EVENT__HAVE_ACCEPT4 /* #undef EVENT__HAVE_ARC4RANDOM / While autotools has: #define EVENT__HAVE_ACCEPT4 1 / #undef EVENT__HAVE_ARC4RANDOM */
fa135bdc	2017-01-30T03:17:41	cmake: Fix checking of enum values from sysctl.h CheckSymbolExists do not do this, so add new CheckConstExists that will use CheckCSourceCompiles() to check this. v2: use set() instead of string(APPEND)
c199df7b	2017-03-06T02:52:46	travis-ci: remove -DEVENT__ENABLE_VERBOSE_DEBUG=ON (too much useless logs) I guess we can revert part of 59649f7 commit, we tried, but it adds more problems than solves. Refs: https://github.com/libevent/libevent/pull/476#issuecomment-284259602
98faf198	2017-03-06T02:46:23	Merge branch 'fix-openssl-linking' * fix-openssl-linking: sample/https-client: use ERR_remove_*state() when we have them Do not check for ERR_remove_thread_state() (do not link ssl into every library) Closes: #476
05318f13	2017-03-05T11:33:44	sample/https-client: use ERR_remove_*state() when we have them ERR_remove_thread_state: - openssl 0.9.8 does not have - openssl 1.0/libressl has - openssl 1.1 deprecates
5d18d879	2017-03-02T21:09:32	Do not check for ERR_remove_thread_state() (do not link ssl into every library) This reverts commit c4e9d9bd662de7f575f2172c160795d452ebe709 ("sample/https-client: check for ERR_remove_thread_state() existence"). Calling AC_SEARCH_LIBS() modifies LIBS - -lcrypto incorrectly ends up in LIBS, and thus linked to by libevent_core.so. Checking for ERR_remove_thread_state should no longer be needed because it was introduced in openssl 1.0.0, and the previous line 0.9.8 had support discontinued at the end of 2015. Fixes: #473
8df81546	2017-03-05T23:10:27	evutil: fix a typo of comment Fixes: #481 Fixes: #480
cc0e04d7	2017-02-21T12:54:02	Fix RPATH for APPLE By setting the CMake minimum version to 3.1 CMake automatically adds the correct magic to make the library relocatable on the filesystem (instead of burning the location of the library at link time into the binary). ex: otool -L bin/http-connect bin/http-connect: @rpath/libevent_extra.2.2.0.dylib (compatibility version 2.2.0, current version 0.0.0) /usr/lib/libz.1.dylib (compatibility version 1.0.0, current version 1.2.8) /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 1238.0.0) Fixes: #468 (cherry-picked)
55cadb2b	2017-02-20T16:55:57	sample/dns-example: fix compiler warning (getopt() returns an int) Signed-off-by: David Disseldorp <ddiss@suse.de> Closes: #449
e139cbac	2017-02-20T14:47:59	sample/https-client: use host SSL certificate store by default Currently a static (Debian) certificate path is used by default, which can be overridden using the -crt parameter. This commit changes the default behaviour such that the openssl default certificate store is used, unless overridden by -crt. Signed-off-by: David Disseldorp <ddiss@suse.de>
ffbce578	2017-01-30T01:58:22	Fix clang 3 (some versions) misbehaviour for __has_attribute() Refs: #447 Clang: https://reviews.llvm.org/rL223468 ("Modify __has_attribute so that it only looks for GNU-style attributes")
d057c45e	2017-01-28T16:42:13	Unbreak build with LibreSSL after openssl 1.1 support added Fixes: 3e9e0a0d46e4 ("Make it build using OpenSSL 1.1.0") Fixes: #445
9d5a4bdc	2017-01-29T23:07:40	Merge branch 'fix-struct-linger' @jbech "Accidentally disabled by 0dda56a due to confusion between struct linger vs. SO_LINGER and #define vs. AC_DEFINE. Try adding synthetic #error test to confirm." * fix-struct-linger: cmake: check for 'struct linger' existence test/bench*: prefix event-config.h macros after 0dda56a48e94 test/bench_httpclient: restore SO_LINGER usage after 0dda56a48e94 Fixes: #444 (original pull-request)
b24a78cc	2017-01-29T23:05:18	cmake: check for 'struct linger' existence
77ec05e5	2017-01-28T16:28:57	test/bench*: prefix event-config.h macros after 0dda56a48e94
d9118c8d	2017-01-28T16:17:07	test/bench_httpclient: restore SO_LINGER usage after 0dda56a48e94
7bc822ca	2017-01-29T21:37:31	Include openssl-compat.h into dist archive Refs: nmathewson/Libevent@140 Refs: http://archives.seul.org/libevent/users/Jan-2017/msg00001.html
fe2c2622	2017-01-29T20:19:06	Make EVENT_STAGE_NAME non cached It did not work before correctly anyway, since cmake do not reset cache entries by default, so that reset to "beta" didn't work. But I don't think that making this variable cached is useful, so let's remove this.
0344edf6	2017-01-29T19:32:10	Do not overwrite version from GIT if it is older then defaults
efcfa046	2017-01-29T18:13:07	Bump version in master to 2.2.0-alpha-dev And according to f8d80a285cf941505277d9060d0357b5b7e543d2 ("Bump version to 2.1.3-alpha-dev") numeric version have 1 at the end, since it means "-dev".
94e7dceb	2017-01-29T17:23:14	Fix -Werror=implicit-fallthrough (fixes gcc-7) Fixes: #447
e7ff4ef2	2017-01-26T02:37:15	Merge branch 'release-2.1.8-stable-pull' * release-2.1.8-stable-pull: Bump version to 2.1.8-stable everywhere VERSION_INFO should me equal to library version Bump whatsnew-2.1 document Update ChangeLog for 2.1.8-stable README: update AUTHORS
9b63b5d4	2017-01-22T18:21:57	Bump version to 2.1.8-stable everywhere P.S. in referenced commit I made a mistake and instead of -rc I wrote -beta. See-also: f4489b8323cda0bf8eed7c1353911a852d7a947c ("Bump version to 2.1.7-beta everywhere")
bed52ff6	2017-01-22T16:38:01	VERSION_INFO should me equal to library version This reverts commit 43243b9d55efcd0e6e2767a5c45bc7da46cff60c (I shouldn't do this).
227bfe56	2017-01-22T16:58:12	Bump whatsnew-2.1 document
d9eb42e0	2017-01-22T16:16:58	Update ChangeLog for 2.1.8-stable $ git log --format=' o %s (%h %cN)' HEAD...refs/tags/release-2.1.7-rc
50824ae2	2017-01-22T16:14:30	README: update AUTHORS $ git log --format='%an' HEAD...refs/tags/release-2.1.6-beta \| sort -u \| xargs -n1 -i bash -c 'fgrep -q "{}" README.md \|\| echo {}'
59e217df	2017-01-22T03:59:02	Merge branch 'automake-tests-parallel-v4' (includes ci bits also) This patch set runs tests in parallel (on travis/appveyor/vagrant), it includes cmake/autotools(automake). It should significantly decrease time that tests tooks on travis-ci (3-4 times lower, right now it is about 14-17 hours - too long!), but not without downsides, now because travis-ci workers has limited resources we will have more timing-related failures, but this is another storry, anyway ~16 hours is not acceptable. Anyway if machine has enough resources it is great to have ability to run tests in parallel (which automake couldn't do before). Changes for common test env: - autotools: before 17m / after 3m - cmake: before 15m / after 87 sec * automake-tests-parallel-v4: Run tests in parallel (they are lightweight), on CI and vagrant boxes travis-ci: install cmake from xenial (CTEST_PARALLEL_LEVEL support) test: register different tests in automake test: run different tests under different options (in a wrapper) automake: do not use serial-tests if parallel-test-harness available Fixes: #439
9c9be399	2017-01-22T03:58:15	vagrant/netbsd: missing libtool
108b5fef	2017-01-20T16:31:17	Run tests in parallel (they are lightweight), on CI and vagrant boxes By default 20 parllel jobs, but one caveat for travis-ci, osx boxes there slower and have less resources then linux (discovered during testing), so limit number of parallel jobs to 4 there, and also install travis_wait, so that travis-ci will not fail the build when there is no output for 10 minutes, since for osx boxes it is very likely. Refs: travis-ci/travis-ci@7202
2a67f4fc	2017-01-20T16:50:04	travis-ci: install cmake from xenial (CTEST_PARALLEL_LEVEL support) cmake supports it since 3.0 I guess. And it must be trusty (not precise), otherwise it fails: dpkg: error processing /var/cache/apt/archives/ifupdown_0.8.10ubuntu1_amd64.deb (--unpack): trying to overwrite '/etc/init.d/networking', which is also in package netbase 4.47ubuntu1 Link: https://travis-ci.org/azat/libevent/jobs/193744023
ea1d30ca	2017-01-20T16:29:19	test: register different tests in automake Before this patch we have one test.sh (well test-script.sh), and tooks very long to run it sequentially, but they are pretty lightweight, so we should run then in parallel.
54d7473f	2017-01-20T16:29:02	test: run different tests under different options (in a wrapper)
a7513cc9	2017-01-15T12:31:27	automake: do not use serial-tests if parallel-test-harness available Revert: 61179dec7716975a151c6dae4524ae504d05780e ("automake: define serial-tests only if automake have this option") Revert: Revert: 44d755e00e249c09914cf35089c16735df82e270 ("test/automake: don't use paralell test harness (since automake 1.12)")
43243b9d	2017-01-20T21:24:54	Bump VERSION_INFO so that libraries symlinks will have the same version Fixes: f4489b8323cda0bf8eed7c1353911a852d7a947c ("Bump version to 2.1.7-beta everywhere")
d6bafbbe	2017-01-20T02:11:35	test/dns: replace servname since solaris does not have "http" Yes, by default solaris (solaris10.dev 5.10 Generic_147148-26 i86pc i386 i86pc) does not have "http" in /etc/services
83e0f43b	2017-01-19T20:53:18	Merge branch 'openssl-filter-fixes-v4' * openssl-filter-fixes-v4: be: fix with filtered bufferevents and connect() without EAGAIN test/ssl: fix bufferevent_getfd() for bufferevent_openssl_filter_new() be_openssl: Fix writing into filted openssl bufferevent after connected test/https: separate cases for https client with filtered openssl bufferevent test/ssl: cover case when we writing to be_openssl after connecting Fixes: #428
9a0a3a3e	2017-01-16T02:31:54	be: fix with filtered bufferevents and connect() without EAGAIN With filtered bufferevents (i.e. not real one, that have socket), we can trigger incorrect callback in this case. Let's look at example with http and bufferevent_openssl_filter_new(): - bev = bufferevent_openssl_filter_new() - http layer trying to connect() to localhost with bev # at this time, bev have writecb/readcb NULL but ev_write/ev_read has # timeout with 45 secs, default HTTP connect timeout - and when connect() retruns without EAGAIN (BSD'ism) we called event_active() before (with EV_WRITE), and this will call ev_write timeout only, while it is more correct to act on bufferevent instead of plain event, so let's trigger EV_WRITE for bufferevent which will do the job (and let's do this deferred). Fixes: http/https_simple # under solaris
09b62013	2016-12-08T02:11:22	test/ssl: fix bufferevent_getfd() for bufferevent_openssl_filter_new()
89396767	2016-12-07T02:53:07	be_openssl: Fix writing into filted openssl bufferevent after connected The main problems was due to when bufferevent_openssl has underlying (i.e. created with bufferevent_openssl_filter_new()) some events was disabled/suspended, while with openssl, READ can require WRITE and vice-versa hence this issues. The BEV_CTRL_GET_FD hunk to fix http subsystem, since it depends from what bufferevent_getfd() returns. Fixes: #428 Fixes: ssl/bufferevent_filter_write_after_connect Fixes: http/https_filter_chunk_out Fixes: da52933550fd4736aa1c213b6de497e2ffc31e34 ("be_openssl: don't call do_write() directly from outbuf_cb")
d77fcea1	2016-12-07T03:07:59	test/https: separate cases for https client with filtered openssl bufferevent - http/https_filter_chunk_out # now hang - http/https_filter_basic # works, since writes only before connect()
d047c241	2016-12-02T18:32:03	test/ssl: cover case when we writing to be_openssl after connecting Right now it fails because of regression for filtered openssl bufferevent, and by it I mean ssl/bufferevent_filter_write_after_connect test, and by fails - hang. Regression-for: da52933550fd4736aa1c213b6de497e2ffc31e34 ("be_openssl: don't call do_write() directly from outbuf_cb")
3c7422fc	2017-01-19T20:51:38	test/thread: netbsd is too slow, increase timeout for conditions_simple
07862531	2017-01-19T20:34:41	test/dns: run async resolving after sync one (to avoid timeouts) If system resolver (sync one) will respond too slow, then we can fail async request and evdns will retransmit tham again, but evdns server will accept that failed requets, so we will have not 2 requests but 4. Reproduced on centos box sometimes.
36da6877	2017-01-18T14:33:56	vagrant/netbsd: more reliable way of installing packages Link: http://ftp.netbsd.org/pub/pkgsrc/current/pkgsrc/doc/pkgsrc.html#installing-binary-packages

80852425

2017-06-30T02:27:08

Document some obvious cases where a function might also return NULL Closes: #525

cd285e42

2017-05-29T22:11:48

Fix event_debug_logging_mask_ exporting on win32

1771ec0f

2017-05-29T22:12:09

Vagrant test

3ac4f176

2017-05-29T22:07:57

Revert "Make event_debug_get_logging_mask_() real symbol (win32)" This reverts commit 5f17c12de021c070501e97268e595da82bf38879.

5f17c12d

2017-05-29T20:46:33

Make event_debug_get_logging_mask_() real symbol (win32) I cannot made it work without this, even though dumpbin shows that that symbol exists in the event_core.dll, event_extra.dll failed to compile: ==> win: Creating library C:/vagrant/.cmake-vagrant/lib/Debug/event_extra.lib and object C:/vagrant/.cmake-vagrant/lib/Debug/event_extra.exp ==> win: http.obj : error LNK2001: unresolved external symbol _event_debug_logging_mask_ [C:\vagrant\.cmake-vagrant\event_extra_shared.vcxproj] ==> win: C:\vagrant\.cmake-vagrant\bin\Debug\event_extra.dll : fatal error LNK1120: 1 unresolved externals [C:\vagrant\.cmake-vagrant\event_extra_shared.vcxproj] ==> win: Done Building Project "C:\vagrant\.cmake-vagrant\event_extra_shared.vcxproj" (default targets) -- FAILED. And dumpbin: sh-4.1$ /cygdrive/c/Program\ Files\ $x86$/Microsoft\ Visual\ Studio\ 12.0/VC/bin/amd64/dumpbin.exe /EXPORTS ./bin/Debug/event_core.dll M 202 C9 00059A3C event_debug_logging_mask_ = _event_debug_logging_mask_

ce3af533

2017-05-29T15:04:50

Fix visibility issues under (mostly on win32) Refs: #511 Fixes: 7182c2f561570cd9ceb704623ebe9ae3608c7b43 ("cmake: build SHARED and STATIC libraries (like autoconf does)")

8b29b136

2017-05-29T16:36:24

cmake: fix building bench/bench_cascade under win32 (missing getopt) Because of typo in cmake, now rewrote to make it less error prone (since even for non-win32 there are variables overlap).

346f8e17

2017-05-29T16:32:30

cmake: avoid requiring cmake 3.1 on win32 for dns-example Avoid using target_sources() from cmake 3.1, just pass needable sources as list.

2773a5ed

2017-05-29T18:48:48

cmake: set CMP0054 to NEW to avoid variables over expansion (since cmake 3.8) ==> win: CMake Warning (dev) at CMakeLists.txt:782 (elseif): ==> win: Policy CMP0054 is not set: Only interpret if() arguments as variables or ==> win: keywords when unquoted. Run "cmake --help-policy CMP0054" for policy ==> win: details. Use the cmake_policy command to set the policy and suppress this ==> win: warning. ==> win: ==> win: Quoted variables like "MSVC" will no longer be dereferenced when the policy ==> win: is set to NEW. Since the policy is not set the OLD behavior will be used. ==> win: This warning is for project developers. Use -Wno-dev to suppress it.

74a2f593

2017-05-29T16:24:48

le-proxy: fix compiling under win32 (because of no "inline" in C) Include <util-internal.h> for regular "inline", since it is used everywhere.

d22e12ee

2017-05-28T16:38:20

http-connect: exit on error

15da23cf

2017-05-01T21:56:27

test: do not use .fieldname in structure initializations (fixes win32) Fixes: #497

8d89c212

2017-05-01T21:55:00

Add missing print-winsock-errors.c into dist archive Refs: #497

00ae45fe

2017-04-25T23:10:13

le-proxy: add missing return statement (UAF) Refs: #501501501501501501501501501

0b4b0efd

2015-04-13T08:54:52

Return from event_del() after the last event callback termination Delete the event from the queue before blocking for the current event callback termination. Ensures that no callback is being executed when event_del() returns, hence making this function a secure mechanism to access data which is handled in the event callack. Fixes: #236 Fixes: #225 Refs: 6b4b77a Fixes: del_wait

5ff83989

2017-04-30T01:11:31

test: cover that after event_del() callback will not be scheduled again Refs: #236 Refs: #225

f988c5ca

2017-04-28T19:36:42

Correct URL for the libevent logo

c6c74ce2

2017-04-20T18:03:50

Explicitly call SSL_clear when reseting the fd. If reconnecting the via BEV_CTRL_SET_FD, bufferevent_openssl.c expects OpenSSL to reuse the configuration state in the SSL object but retain connection state. This corresponds to the SSL_clear API. The code currently only calls SSL_set_connect_state or SSL_set_accept_state. Due to a quirk in OpenSSL, doing this causes the handshake to implicitly SSL_clear the next time it is entered. However, this, in the intervening time, leaves the SSL object in an odd state as the connection state has not been dropped yet. This behavior also does not appear to be documented by OpenSSL. Instead, call SSL_clear explicitly: https://www.openssl.org/docs/manmaster/man3/SSL_clear.html

92cc0b9c

2017-04-13T14:58:13

bufferevent: refactor to use type check macros

266f43af

2017-03-27T15:50:23

Fix arc4random_addrandom() detecting and fallback (regression) But this is kind of hot-fix, we definitelly need more sane arc4random compat layer. Fixes: #488 Introduced-in: 6541168 ("Detect arch4random_addrandom() existence")

177e2171

2017-03-24T16:49:59

Make event_count macros cleaner Fixes: #489

2e52bace

2017-03-16T23:32:43

fix evutil_make_internal_pipe_'s comment that fd[0] for read and f[1] for write

b1e8a413

2017-03-14T13:33:31

cmake: use APPEND during exporting targets (for old cmake) On centos with cmake 2.8.12.2: CMake Error at cmake/AddEventLibrary.cmake:92 (export): export called with target "event_extra_shared" which requires target "event_core_shared" that is not in the export list. If the required target is not easy to reference in this call, consider using the APPEND option with multiple separate calls. But on newer cmake I guess everything is ok. Fixes: 7182c2f561570cd9ceb704623ebe9ae3608c7b43 ("cmake: build SHARED and STATIC libraries (like autoconf does)")

d84f0205

2017-03-14T13:31:02

Merge branch 'win32-fixes' This patchset fixes win32 builds after some previous patches (referenced in particular commits), and also removes some quirks for win32. * win32-fixes: test: do not return void log/win32: fix exporting extern variable log-internal: missing extern "C" log: remove USE_GLOBAL_FOR_DEBUG_LOGGING cmake: Export missing symbols for win32 cmake: eliminate EVENT_BUILDING_REGRESS_TEST, since we link with shared libs test: windows doesn't have WNOWAIT cmake: clean not used #defines from event-config.h cmake: add <pthread.h> only for non-win32

8acfb0cd

2017-03-14T13:21:16

test: do not return void

92cf234b

2017-03-14T00:33:26

log/win32: fix exporting extern variable ==> win: C:\vagrant\log.c(73): error C2370: 'event_debug_logging_mask_' : redefinition; different storage class [C:\vagrant\.cmake-vagrant\event_core_shared.vcxproj]

a9c2116f

2017-03-14T01:06:10

log-internal: missing extern "C"

49dcf4d5

2017-03-13T22:33:38

log: remove USE_GLOBAL_FOR_DEBUG_LOGGING Fixes: 9081b66c9c1446b11c0052bade799fafb460efed ("Export symbols for -fvisibility=hidden (under cmake)")

48ead58e

2017-03-14T00:46:47

cmake: Export missing symbols for win32

30f2a969

2017-03-14T00:07:17

cmake: eliminate EVENT_BUILDING_REGRESS_TEST, since we link with shared libs Before 7182c2f561570cd9ceb704623ebe9ae3608c7b43 ("cmake: build SHARED and STATIC libraries (like autoconf does)") it links with *.c.

a4b1828d

2017-03-13T23:52:15

test: windows doesn't have WNOWAIT Fixes: 66a4eb0c3ae3b1f22b084b2d3aeb5c872f37efbd ("Check for WNOWAIT in waitpid() in runtime (not in cmake/configure)")

d5e10870

2017-03-13T21:54:33

cmake: clean not used #defines from event-config.h - _LARGE_FILES: we have it in evconfig-private.h.in, drop from event-config.h - EVENT___FILE_OFFSET_BITS: remove it for now - EVENT__const

2d2299cf

2017-03-13T22:02:45

cmake: add <pthread.h> only for non-win32 Fixes: b2b4b4d74e78a3e4fe7a74224d1e6aada5bde351 ("cmake: add <pthread.h> into CMAKE_REQUIRED_INCLUDES for sizeof(pthread_t)")

882f537c

2017-03-13T21:40:52

cmake: fix pkgconfig generation (copy-paste typo)

489991a2

2017-03-13T21:05:32

Merge branch 'cmake-missing-bits' This patchset adds next missing things (in compare to autotools): - pkgconfig - event_pthreads/event_openssl - compile shared/static libraries always And some fixes, because it will not build after fixing other things: - export missing symbols for cmake (-fvisibility=hidden) * cmake-missing-bits: cmake: support visibility for AppleClang too cmake: fix export absolute path and relative path and cleanup a bit cmake: generate and install pkgconfig files cmake: build SHARED and STATIC libraries (like autoconf does) cmake: add missing event_openssl/event_pthreads libraries Export symbols for -fvisibility=hidden (under cmake) Refs: #246

805e1aa8

2017-03-13T15:49:11

cmake: support visibility for AppleClang too Defaults apple linker behaviour is -two_levelnamespace, and you cannot use "-undefined suppress" with it, so let's link non event_core with event_core library to avoid undefined symbols.

fd5a3fc0

2017-03-13T03:18:02

cmake: fix export absolute path and relative path and cleanup a bit Remove next vars, since I don't think that somebody want to change it: - bin - lib - include And fix exports problem: CMake Error: INSTALL(EXPORT "LibeventTargets") given absolute DESTINATION "/usr/lib/cmake/libevent" but the export references an installation of target "event_core_static" which has relative DESTINATION "lib".

86c73575

2017-03-13T02:26:39

cmake: generate and install pkgconfig files Fixes: #443

7182c2f5

2017-03-12T23:31:59

cmake: build SHARED and STATIC libraries (like autoconf does) Since they are useful for debugging, and if autotools build them then cmamke has to do this too, to make migration more simple. And now: - tests: uses shared libraries (since this is upstreams one) - other binaries: uses static libraries This removes next private config: - EVENT__NEED_DLLIMPORT

72ef9d16

2016-11-07T00:46:45

cmake: add missing event_openssl/event_pthreads libraries This will remove openssl requirement if you don't use it (i.e. if you not link with openssl_pthreads). Plus it fixes some linking dependencies: - libm required only for test-ratelim And fix some coding style alignment issues. Refs: #246

9081b66c

2017-03-12T20:50:35

Export symbols for -fvisibility=hidden (under cmake) Fixes: #442

83b15847

2017-03-12T13:12:53

travis-ci: do not run coverage on osx First of all we don't really need two of them, and also with apple-clang it will fail (because of our checks in cmake), so just exclude it from the build matrix.

7004ee8e

2017-03-09T09:32:43

travis-ci: use travis_retry to reduce number of problems with flacky tests

e01b9937

2017-03-08T23:48:30

appveyor: remove -DEVENT__ENABLE_VERBOSE_DEBUG=ON build See-also: c199df7bc78824ff579ff34c5f9f922034e8fa31 ("travis-ci: remove -DEVENT__ENABLE_VERBOSE_DEBUG=ON (too much useless logs)")

65870949

2017-03-12T01:00:41

Fix detection of timerfd_create() in CMake.

819d0493

2017-03-08T23:15:33

Do not add epoll_sub (syscall wrappers) for epoll in cmake Nowadays mostly all supported OS'es has this wrappers, and some of them (like SmartOS) has wrappers but doesn't have __NR_epoll* defines for syscall numbers, so just drop them (instead of adding yet another check int cmake like autotools has, since this will break building in cross-compile environment). Also one minor note, configure doesn't added epoll_sub.c either, since it check epoll_create() in runtime. And I tested it in SmartOS and it even works (`make verify`). Fixes: #463

5aade2d3

2017-03-08T17:58:26

Merge branch 'cmake-configure-fixes-v2' Fixes in cmake, to make it more like configure and support some cross-compiling. * cmake-configure-fixes-v2: cmake: fix extracting of the version from git (check for number of matches) Detect arch4random_addrandom() existence Use off_t instead of ev_off_t for sendfile() (fixes android build) cmake: detect _GNU_SOURCE not by __GNU_LIBRARY__ only (fallback to _GNU_SOURCE) Check for WNOWAIT in waitpid() in runtime (not in cmake/configure) cmake: add <pthread.h> into CMAKE_REQUIRED_INCLUDES for sizeof(pthread_t) cmake: fix values for #cmakedefine cmake: drop duplicates from event-config template cmake: add value for the #cmakedefine macros (like autoconf) cmake: Fix checking of enum values from sysctl.h

d65e3e1d

2017-03-06T14:48:56

cmake: fix extracting of the version from git (check for number of matches)

6541168d

2017-03-06T00:55:16

Detect arch4random_addrandom() existence Refs: #370 Refs: #475

f479a008

2017-03-06T00:45:09

Use off_t instead of ev_off_t for sendfile() (fixes android build) Refs: #475

71a56bd3

2017-03-06T00:30:21

cmake: detect _GNU_SOURCE not by __GNU_LIBRARY__ only (fallback to _GNU_SOURCE) Fixes pipe2() on Android-9/armv7-a Refs: #475

66a4eb0c

2017-03-06T00:05:50

Check for WNOWAIT in waitpid() in runtime (not in cmake/configure) Because checking in cmake breaks cross-compiling. Introduced-in: 43eb56c7c738e3642f0981e3dd6ab9e082eec798. Fixes: #482 Fixes: #462 Refs: #475 v2: use waitid() with WNOWAIT v3: use WNOWAIT only if it available in waitpid(), because not all netbsd supports it

b2b4b4d7

2017-03-05T22:07:23

cmake: add <pthread.h> into CMAKE_REQUIRED_INCLUDES for sizeof(pthread_t) Fixes: #475

7f945b2d

2017-03-05T21:58:20

cmake: fix values for #cmakedefine

85272602

2017-03-05T21:57:45

cmake: drop duplicates from event-config template

336ab816

2017-03-05T21:48:56

cmake: add value for the #cmakedefine macros (like autoconf) Before this patch for cmake we had: #define EVENT__HAVE_ACCEPT4 /* #undef EVENT__HAVE_ARC4RANDOM */ While autotools has: #define EVENT__HAVE_ACCEPT4 1 /* #undef EVENT__HAVE_ARC4RANDOM */

fa135bdc

2017-01-30T03:17:41

cmake: Fix checking of enum values from sysctl.h CheckSymbolExists do not do this, so add new CheckConstExists that will use CheckCSourceCompiles() to check this. v2: use set() instead of string(APPEND)

c199df7b

2017-03-06T02:52:46

travis-ci: remove -DEVENT__ENABLE_VERBOSE_DEBUG=ON (too much useless logs) I guess we can revert part of 59649f7 commit, we tried, but it adds more problems than solves. Refs: https://github.com/libevent/libevent/pull/476#issuecomment-284259602

98faf198

2017-03-06T02:46:23

Merge branch 'fix-openssl-linking' * fix-openssl-linking: sample/https-client: use ERR_remove_*state() when we have them Do not check for ERR_remove_thread_state() (do not link ssl into every library) Closes: #476

05318f13

2017-03-05T11:33:44

sample/https-client: use ERR_remove_*state() when we have them ERR_remove_thread_state: - openssl 0.9.8 does not have - openssl 1.0/libressl has - openssl 1.1 deprecates

5d18d879

2017-03-02T21:09:32

Do not check for ERR_remove_thread_state() (do not link ssl into every library) This reverts commit c4e9d9bd662de7f575f2172c160795d452ebe709 ("sample/https-client: check for ERR_remove_thread_state() existence"). Calling AC_SEARCH_LIBS() modifies LIBS - -lcrypto incorrectly ends up in LIBS, and thus linked to by libevent_core.so. Checking for ERR_remove_thread_state should no longer be needed because it was introduced in openssl 1.0.0, and the previous line 0.9.8 had support discontinued at the end of 2015. Fixes: #473

8df81546

2017-03-05T23:10:27

evutil: fix a typo of comment Fixes: #481 Fixes: #480

cc0e04d7

2017-02-21T12:54:02

Fix RPATH for APPLE By setting the CMake minimum version to 3.1 CMake automatically adds the correct magic to make the library relocatable on the filesystem (instead of burning the location of the library at link time into the binary). ex: otool -L bin/http-connect bin/http-connect: @rpath/libevent_extra.2.2.0.dylib (compatibility version 2.2.0, current version 0.0.0) /usr/lib/libz.1.dylib (compatibility version 1.0.0, current version 1.2.8) /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 1238.0.0) Fixes: #468 (cherry-picked)

55cadb2b

2017-02-20T16:55:57

sample/dns-example: fix compiler warning (getopt() returns an int) Signed-off-by: David Disseldorp <ddiss@suse.de> Closes: #449

e139cbac

2017-02-20T14:47:59

sample/https-client: use host SSL certificate store by default Currently a static (Debian) certificate path is used by default, which can be overridden using the -crt parameter. This commit changes the default behaviour such that the openssl default certificate store is used, unless overridden by -crt. Signed-off-by: David Disseldorp <ddiss@suse.de>

ffbce578

2017-01-30T01:58:22

Fix clang 3 (some versions) misbehaviour for __has_attribute() Refs: #447 Clang: https://reviews.llvm.org/rL223468 ("Modify __has_attribute so that it only looks for GNU-style attributes")

d057c45e

2017-01-28T16:42:13

Unbreak build with LibreSSL after openssl 1.1 support added Fixes: 3e9e0a0d46e4 ("Make it build using OpenSSL 1.1.0") Fixes: #445

9d5a4bdc

2017-01-29T23:07:40

Merge branch 'fix-struct-linger' @jbech "Accidentally disabled by 0dda56a due to confusion between struct linger vs. SO_LINGER and #define vs. AC_DEFINE. Try adding synthetic #error test to confirm." * fix-struct-linger: cmake: check for 'struct linger' existence test/bench*: prefix event-config.h macros after 0dda56a48e94 test/bench_httpclient: restore SO_LINGER usage after 0dda56a48e94 Fixes: #444 (original pull-request)

b24a78cc

2017-01-29T23:05:18

cmake: check for 'struct linger' existence

77ec05e5

2017-01-28T16:28:57

test/bench*: prefix event-config.h macros after 0dda56a48e94

d9118c8d

2017-01-28T16:17:07

test/bench_httpclient: restore SO_LINGER usage after 0dda56a48e94

7bc822ca

2017-01-29T21:37:31

Include openssl-compat.h into dist archive Refs: nmathewson/Libevent@140 Refs: http://archives.seul.org/libevent/users/Jan-2017/msg00001.html

fe2c2622

2017-01-29T20:19:06

Make EVENT_STAGE_NAME non cached It did not work before correctly anyway, since cmake do not reset cache entries by default, so that reset to "beta" didn't work. But I don't think that making this variable cached is useful, so let's remove this.

0344edf6

2017-01-29T19:32:10

Do not overwrite version from GIT if it is older then defaults

efcfa046

2017-01-29T18:13:07

Bump version in master to 2.2.0-alpha-dev And according to f8d80a285cf941505277d9060d0357b5b7e543d2 ("Bump version to 2.1.3-alpha-dev") numeric version have 1 at the end, since it means "-dev".

94e7dceb

2017-01-29T17:23:14

Fix -Werror=implicit-fallthrough (fixes gcc-7) Fixes: #447

e7ff4ef2

2017-01-26T02:37:15

Merge branch 'release-2.1.8-stable-pull' * release-2.1.8-stable-pull: Bump version to 2.1.8-stable everywhere VERSION_INFO should me equal to library version Bump whatsnew-2.1 document Update ChangeLog for 2.1.8-stable README: update AUTHORS

9b63b5d4

2017-01-22T18:21:57

Bump version to 2.1.8-stable everywhere P.S. in referenced commit I made a mistake and instead of *-rc* I wrote *-beta*. See-also: f4489b8323cda0bf8eed7c1353911a852d7a947c ("Bump version to 2.1.7-beta everywhere")

bed52ff6

2017-01-22T16:38:01

VERSION_INFO should me equal to library version This reverts commit 43243b9d55efcd0e6e2767a5c45bc7da46cff60c (I shouldn't do this).

227bfe56

2017-01-22T16:58:12

Bump whatsnew-2.1 document

d9eb42e0

2017-01-22T16:16:58

Update ChangeLog for 2.1.8-stable $ git log --format=' o %s (%h %cN)' HEAD...refs/tags/release-2.1.7-rc

50824ae2

2017-01-22T16:14:30

README: update AUTHORS $ git log --format='%an' HEAD...refs/tags/release-2.1.6-beta | sort -u | xargs -n1 -i bash -c 'fgrep -q "{}" README.md || echo {}'

59e217df

2017-01-22T03:59:02

Merge branch 'automake-tests-parallel-v4' (*includes ci bits also*) This patch set runs tests in parallel (on travis/appveyor/vagrant), it includes cmake/autotools(automake). It should significantly decrease time that tests tooks on travis-ci (3-4 times lower, right now it is about 14-17 hours - too long!), but not without downsides, now because travis-ci workers has limited resources we will have more timing-related failures, but this is another storry, anyway ~16 hours is not acceptable. Anyway if machine has enough resources it is great to have ability to run tests in parallel (which automake couldn't do before). Changes for common test env: - autotools: before 17m / after 3m - cmake: before 15m / after 87 sec * automake-tests-parallel-v4: Run tests in parallel (they are lightweight), on CI and vagrant boxes travis-ci: install cmake from xenial (CTEST_PARALLEL_LEVEL support) test: register different tests in automake test: run different tests under different options (in a wrapper) automake: do not use serial-tests if parallel-test-harness available Fixes: #439

9c9be399

2017-01-22T03:58:15

vagrant/netbsd: missing libtool

108b5fef

2017-01-20T16:31:17

Run tests in parallel (they are lightweight), on CI and vagrant boxes By default 20 parllel jobs, but one caveat for travis-ci, osx boxes there slower and have less resources then linux (discovered during testing), so limit number of parallel jobs to 4 there, and also install travis_wait, so that travis-ci will not fail the build when there is no output for 10 minutes, since for osx boxes it is very likely. Refs: travis-ci/travis-ci@7202

2a67f4fc

2017-01-20T16:50:04

travis-ci: install cmake from xenial (CTEST_PARALLEL_LEVEL support) cmake supports it since 3.0 I guess. And it must be trusty (not precise), otherwise it fails: dpkg: error processing /var/cache/apt/archives/ifupdown_0.8.10ubuntu1_amd64.deb (--unpack): trying to overwrite '/etc/init.d/networking', which is also in package netbase 4.47ubuntu1 Link: https://travis-ci.org/azat/libevent/jobs/193744023

ea1d30ca

2017-01-20T16:29:19

test: register different tests in automake Before this patch we have one test.sh (well test-script.sh), and tooks very long to run it sequentially, but they are pretty lightweight, so we should run then in parallel.

54d7473f

2017-01-20T16:29:02

test: run different tests under different options (in a wrapper)

a7513cc9

2017-01-15T12:31:27

automake: do not use serial-tests if parallel-test-harness available Revert: 61179dec7716975a151c6dae4524ae504d05780e ("automake: define serial-tests only if automake have this option") Revert: Revert: 44d755e00e249c09914cf35089c16735df82e270 ("test/automake: don't use paralell test harness (since automake 1.12)")

43243b9d

2017-01-20T21:24:54

Bump VERSION_INFO so that libraries symlinks will have the same version Fixes: f4489b8323cda0bf8eed7c1353911a852d7a947c ("Bump version to 2.1.7-beta everywhere")

d6bafbbe

2017-01-20T02:11:35

test/dns: replace servname since solaris does not have "http" Yes, by default solaris (solaris10.dev 5.10 Generic_147148-26 i86pc i386 i86pc) does not have "http" in /etc/services

83e0f43b

2017-01-19T20:53:18

Merge branch 'openssl-filter-fixes-v4' * openssl-filter-fixes-v4: be: fix with filtered bufferevents and connect() without EAGAIN test/ssl: fix bufferevent_getfd() for bufferevent_openssl_filter_new() be_openssl: Fix writing into filted openssl bufferevent after connected test/https: separate cases for https client with filtered openssl bufferevent test/ssl: cover case when we writing to be_openssl after connecting Fixes: #428

9a0a3a3e

2017-01-16T02:31:54

be: fix with filtered bufferevents and connect() without EAGAIN With filtered bufferevents (i.e. not real one, that have socket), we can trigger incorrect callback in this case. Let's look at example with http and bufferevent_openssl_filter_new(): - bev = bufferevent_openssl_filter_new() - http layer trying to connect() to localhost with bev # at this time, bev have writecb/readcb NULL but ev_write/ev_read has # timeout with 45 secs, default HTTP connect timeout - and when connect() retruns without EAGAIN (BSD'ism) we called event_active() before (with EV_WRITE), and this will call ev_write timeout only, while it is more correct to act on bufferevent instead of plain event, so let's trigger EV_WRITE for bufferevent which will do the job (and let's do this deferred). Fixes: http/https_simple # under solaris

09b62013

2016-12-08T02:11:22

test/ssl: fix bufferevent_getfd() for bufferevent_openssl_filter_new()

89396767

2016-12-07T02:53:07

be_openssl: Fix writing into filted openssl bufferevent after connected The main problems was due to when bufferevent_openssl has underlying (i.e. created with bufferevent_openssl_filter_new()) some events was disabled/suspended, while with openssl, READ can require WRITE and vice-versa hence this issues. The BEV_CTRL_GET_FD hunk to fix http subsystem, since it depends from what bufferevent_getfd() returns. Fixes: #428 Fixes: ssl/bufferevent_filter_write_after_connect Fixes: http/https_filter_chunk_out Fixes: da52933550fd4736aa1c213b6de497e2ffc31e34 ("be_openssl: don't call do_write() directly from outbuf_cb")

d77fcea1

2016-12-07T03:07:59

test/https: separate cases for https client with filtered openssl bufferevent - http/https_filter_chunk_out # now hang - http/https_filter_basic # works, since writes only before connect()

d047c241

2016-12-02T18:32:03

test/ssl: cover case when we writing to be_openssl after connecting Right now it fails because of regression for filtered openssl bufferevent, and by it I mean ssl/bufferevent_filter_write_after_connect test, and by fails - hang. Regression-for: da52933550fd4736aa1c213b6de497e2ffc31e34 ("be_openssl: don't call do_write() directly from outbuf_cb")

3c7422fc

2017-01-19T20:51:38

test/thread: netbsd is too slow, increase timeout for conditions_simple

07862531

2017-01-19T20:34:41

test/dns: run async resolving after sync one (to avoid timeouts) If system resolver (sync one) will respond too slow, then we can fail async request and evdns will retransmit tham again, but evdns server will accept that failed requets, so we will have not 2 requests but 4. Reproduced on centos box sometimes.

36da6877

2017-01-18T14:33:56

vagrant/netbsd: more reliable way of installing packages Link: http://ftp.netbsd.org/pub/pkgsrc/current/pkgsrc/doc/pkgsrc.html#installing-binary-packages

kc3-lang/libevent

Log