[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]
[Libevent-users] Please try out this source distribution before it becomes 2.0.11-stable?
- To: libevent-users@xxxxxxxxxxxxx
- Subject: [Libevent-users] Please try out this source distribution before it becomes 2.0.11-stable?
- From: Nick Mathewson <nickm@xxxxxxxxxxxxx>
- Date: Sat, 23 Apr 2011 03:22:16 -0400
- Delivered-to: archiver@xxxxxxxx
- Delivered-to: libevent-users-outgoing@xxxxxxxx
- Delivered-to: libevent-users@xxxxxxxx
- Delivery-date: Sat, 23 Apr 2011 03:22:25 -0400
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:sender:date:x-google-sender-auth :message-id:subject:from:to:content-type; bh=ZciJ5+xAol/H9dervIkTf5q5M/BysaEzEMATkdNlOV8=; b=dPlTCljAX/ePR5rAOnbYBFZUGsDs0hHQsII/RGSB4PIKeIQqgrDHjVG3UFyznjO6sL dhHocPRiZvUKl97ukN5ZCWRnEqeipFX58t2IBfG92RO+TpbIMRULKuGAXxQEhfr4ekmi Q9vKnw2S2aRw9Q4H8v/gZ8KSB2bj/GaSP+PQg=
- Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:date:x-google-sender-auth:message-id:subject :from:to:content-type; b=sjkt98V1e06y+O43tP6beDWch9cP2W+g/YjIrZb6WkwSC93YvBqlQaEBjYCKc7NX1M aOhrmOOTUMaw3ImR3GiB6CT1G3zrCARYymNExOXQJCNscKdFkL/OfLoaWxD0UqKmnhG9 Oq9bYVHMdcAruo/Dke/GX19TjlrPqsNreSReQ=
- Reply-to: libevent-users@xxxxxxxxxxxxx
- Sender: owner-libevent-users@xxxxxxxxxxxxx
Hi, all!
I'm hoping to put out a bugfix release on the libevent 2.0 series this
coming week. With that in mind, I've put up a "dev" release that
contains almost exactly what I'm hoping to put into the stable
release. If there are no complaints, I'll update the changelog,
documentation, and version number, and put this code out as
2.0.11-stable.
http://www.wangafu.net/~nickm/volatile/libevent-2.0.10-stable-dev.tar.gz
This is NOT the real Libevent 2.0.11-stable release! Please do not
bake this into distributions, use it in production binaries, etc. Of
course, I can't stop you, but if you do, I'll be sad. Please wait for
the real thing to come out. This version is just for testing.
Since 2.0 is stable, this is going to be a bugfix release. I wish I
could say it fixed all the bugs in 2.0.10-stable, but there are still
some recalcitrant, tricky, or under-documented ones left. So for now,
I'm not looking for bugs that exist in 2.0.10-stable and exist here
too... The that bugs I *do* most want to know about by next week are
regressions against 2.0.10-stable?
**That is, is there anything that works with 2.0.10-stable, but
breaks with this code?**
Sorry for all the dire warnings, but I really want to make sure that
each stable release is indeed a step ahead. I can only do that with
your help. Please give it a whirl and let me know if it breaks
anything for you that used to work with 2.0.10-stable.
Here's a draft changelog:
=====
BUGFIXES:
o Fix evport handling of POLLHUP and POLLERR (b42ce4b)
o Fix compilation on Windows with NDEBUG (cb8059d)
o Check for POLLERR, POLLHUP and POLLNVAL for Solaris event ports
(0144886 Trond Norbye)
o Detect and handle more allocation failures. (666b096 Jardel Weyrich)
o Use event_err() only if the failure is truly unrecoverable.
(3f8d22a Jardel Weyrich)
o Handle resize failures in the select backend better. (83e805a)
o Correctly free selectop fields when select_resize fails in
select_init (0c0ec0b)
o Make --enable-gcc-warnings a no-op if not using gcc (3267703)
o Fix a type error in our (unused) arc4random_stir() (f736198)
o Correctly detect and stop non-chunked http requests when the body
is too long (63a715e)
o Have event_base_gettimeofday_cached() always return wall-clock time (a459ef7)
o Workaround for http crash bug 3078187 (5dc5662 Tomash Brechko)
o Fix incorrect assertions and possible use-after-free in
evrpc_free() (4b8f02f Christophe Fillot)
o Reset outgoing http connection when read data in idle state.
(272823f Tomash Brechko)
o Fix subtle recursion in evhttp_connection_cb_cleanup(). (218cf19
Tomash Brechko)
o Fix the case when failed evhttp_make_request() leaved request in
the queue. (0d6622e Tomash Brechko)
o Fix a crash bug in evdns server circular list code (00e91b3)
o Handle calloc failure in evdns. (Found by Dave Hart) (364291e)
o Fix a memory leak on win32 socket->event map. (b4f89f0)
o Add a forgotten NULL check to evhttp_parse_headers (12311ff Sebastian Hahn)
o Fix possible NULL-deref in evdns_cancel_request (5208544 Sebastian Hahn)
PORTABILITY:
o Fall back to sscanf if we have no other way to implement strtoll (453317b)
o Build correctly on platforms without sockaddr_storage (9184563)
o Try to build correctly on platforms with no IPv6 support (713c254)
o Build on systems without AI_PASSIVE (cb92113)
o Fix http unit test on non-windows platforms without getaddrinfo (6092f12)
o Do not check for gethostbyname_r versions if we have getaddrinfo (c1260b0)
o Include arpa/inet.h as needed on HPUX (10c834c Harlan Stenn)
o Include util-internal.h as needed to build on platforms with no
sockaddr_storage (bbf5515 Harlan Stenn)
o Check for getservbyname even if not on win32. (af08a94 Harlan Stenn)
o Add -D_OSF_SOURCE to fix hpux builds (0b33479 Harlan Stenn)
o Check for allocation failures in apply_socktype_protocol_hack (637d17a)
o Fix the check for multicast or broadcast addresses in
evutil_check_interfaces (1a21d7b)
o Avoid a free(NULL) if out-of-memory in evdns_getaddrinfo. Found by
Dave Hart (3417f68)
DEFENSIVE PROGRAMMING:
o Add compile-time check for AF_UNSPEC==PF_UNSPEC (3c8f4e7)
BUGS IN TESTS:
o Fix test.sh output on solaris (b4f89b6 Dave Hart)
o Make test-eof fail with a timeout if we never get an eof. (05a2c22
Harlan Stenn)
o Use %s with printf in test.sh (039b9bd)
o Add an assert to appease clang's static analyzer (b0ff7eb Sebastian Hahn)
o Add a forgotten return value check in the unit tests (3819b62 Sebastian Hahn)
o Actually send NULL request in http_bad_request_test (b693c32 Sebastian Hahn)
o add some (void) casts for unused variables (65707d7 Sebastian Hahn)
o Refactor test_getaddrinfo_async_cancel_stress() (48c44a6 Sebastian Hahn)
o Be nice and "handle" error return values in sample code (4bac793
Sebastian Hahn)
o Check return value of evbuffer_add_cb in tests (93a1abb Sebastian Hahn)
o Remote some dead code from dns-example.c (744c745 Sebastian Hahn)
o Zero a struct sockaddr_in before using it (646f9fe Sebastian Hahn)
BUILD FIXES:
o Fix warnings about AC_LANG_PROGRAM usage (f663112 Sebastian Hahn)
o Skip check for zlib if we have no zlib.h (a317c06 Harlan Stenn)
o Fix autoconf bracket issues; make check for getaddrinfo include
netdb.h (833e5e9 Harlan Stenn)
o Correct an AM_CFLAGS to an AM_CPPFLAGS in test/Makefile.am (9c469db
Dave Hart)
o Fix make distcheck & installation of libevent 1 headers (b5a1f9f Dave Hart)
o Fix compilation under LLVM/clang with --enable-gcc-warnings
(ad9ff58 Sebastian Hahn)
FEATURES:
o Make URI parser able to tolerate nonconformant URIs. (95060b5)
DOCUMENTATION:
o Clarify event_set_mem_functions doc (926f816)
o Correct evhttp_del_accept_socket documentation on whether socket is
closed (f665924)
o fix spelling mistake in whatsnew-2.0.txt (deb2f73)
o Fix sample/http-server ipv6 fixes (eb692be)
o Comment internal headers used in sample code. (4eb281c)
o Be explicit about how long event loops run in event.h documentation (f95bafb)
o Add comment to configure.in to explain gc-sections test logic (c621359)
o Fix a couple of memory leaks in samples/http-server.c. Found by
Dave Hart. (2e9f665)
BUILD IMPROVEMENTS:
o Use the gcc -ffunction-segments feature to allow gc when linking
with static libevent (0965c56 Dave Hart)
o Add configure options to disable installation, regression tests
(49e9bb7 Dave Hart)
cheers, and enjoy your weekend,
---
Nick
***********************************************************************
To unsubscribe, send an e-mail to majordomo@xxxxxxxxxxxxx with
unsubscribe libevent-users in the body.