|
3e9aad72
|
2015-01-05T09:34:32
|
|
update changelog
|
|
|
2ff3be6e
|
2015-01-05T09:34:01
|
|
Merge branch '14_cve_2014_6272_v2' into patches-1.4
|
|
|
4228f769
|
2015-01-05T09:07:15
|
|
Increment version for 1.4.15-stable
|
|
|
76f510e3
|
2015-01-05T08:37:02
|
|
Update 1.4 changelog
|
|
|
7b21c4ea
|
2014-01-05T08:30:46
|
|
Fix CVE-2014-6272 in Libevent 1.4
For this fix, we need to make sure that passing too-large inputs to
the evbuffer functions can't make us do bad things with the heap.
|
|
|
10d228c9
|
2014-11-30T21:04:43
|
|
Merge remote-tracking branch 'sourceforge/patches-1.4' into patches-1.4
|
|
|
867f3b38
|
2014-11-30T19:38:54
|
|
Note the progress for the 14 changelog
|
|
|
641d0d88
|
2014-11-30T19:33:03
|
|
Changelog for 1.4.15-stable.
|
|
|
2d8cf0b7
|
2011-04-21T18:16:26
|
|
Defensive programming to prevent (hopefully impossible) stack-stomping
|
|
353b4ac6
|
2011-01-02T10:17:15
|
|
Check for POLLERR, POLLHUP and POLLNVAL for Solaris event ports
|
|
|
e50ba5b6
|
2010-11-19T12:37:49
|
|
Fix a bug that could allow dns requests with duplicate tx ids
Spotted by Marko Kreen
|
|
|
1d6e30e6
|
2010-11-19T12:35:47
|
|
Avoid truncating huge values for content-length
|
|
|
7cf794b6
|
2010-09-03T15:21:09
|
|
Take generated files out of git; add correct m4 magic for libtool to auto* files
|
|
|
823d9be5
|
2010-09-03T15:13:57
|
|
Prefer autoregen -ivf to manual autogen.sh
Suggested by Ralph Castain
|
|
53c47c2e
|
2010-08-30T11:08:59
|
|
Fix compilation on Visual Studio 2010
|
|
|
f0ff7659
|
2010-08-26T15:09:58
|
|
Fix an issue with forking and signal socketpairs in select/poll backends
Nicholas Marriott identifies an issue where a signal socketpair
doesn't get recreated if the event backend doesn't set event_reinit.
|
|
95b71d0c
|
2010-08-24T12:44:33
|
|
Backport doc fix for evhttp_bind_socket.
|
|
|
e0e6958a
|
2010-08-19T14:00:06
|
|
Avoid deadlock when activating signals.
Fixes bug 3048812.
Based on patch by Nicholas Marriott.
|
|
|
1c25b07e
|
2010-08-19T09:38:44
|
|
Fix kquue correctness test on x84_64
Apparently, in our configure.in check for a working kqueue, we were
leaving some fields unset that seemed to irritate 64-bit kqueue a lot.
Found by Christopher Layne
|
|
|
44f04a2b
|
2010-08-19T09:36:33
|
|
Replace unused case of FD_CLOSEONEXEC with a proper null statement.
|
|
9985231e
|
2010-08-18T11:51:30
|
|
set close-on-exec bit for filedescriptors created by dns subsystem
|
|
|
8a5ebd38
|
2010-08-13T12:12:52
|
|
Backport and tweak the LICENSE file for 1.4
|
|
|
b3d03822
|
2010-07-13T11:06:08
|
|
Pass flags to fcntl(F_SETFL) as int, not long
Everybody but Linux documents this as taking an int, and Linux is
very tolerant of getting an int instead. If it weren't, everybody
doing fcntl(fd,F_SETFL,O_NONBLOCK) would break, since the glibc
headers define O_NONBLOCK as an int literal.
|
|
|
eb8122f2
|
2010-06-19T11:56:12
|
|
Increment the version to 1.4.14b; add a changelog for our VERSION_INFO snafu on 1.4.14
|
|
|
fd32d368
|
2010-06-19T11:51:47
|
|
Set the VERSION_INFO correctly for 1.4.14
I've also added a long comment explaining our history of
VERSION_INFO as relevant for the 1.4.x series.
|
|
|
b3360c16
|
2010-05-27T01:30:23
|
|
Increment Libevent version to 1.4.14-stable
|
|
|
6224ff87
|
2010-05-27T01:11:55
|
|
Changelog for 1.4.14-stable
|
|
|
6cbea13b
|
2010-05-10T19:58:17
|
|
Define _REENTRANT as needed on Solaris, elsewhere
It turns out that _REENTRANT isn't only needed to make certain
functions visible; we also need it to make pthreads work properly some
places (like Solaris, where forgetting _REENTRANT basically means that
all threads are sharing the same errno).
Unlike in 2.0, we can't use the PTHREAD_CFLAGS variable to see what
flags to use, since we aren't using pthreads in 1.4. Instead, we just
check the host type explicitly, like acx_pthreads does.
|
|
|
906d573b
|
2010-05-10T19:57:13
|
|
Fix compile warning in http.c
|
|
29d7b328
|
2010-04-27T22:25:59
|
|
Do not abort HTTP requests missing a reason string.
Some (arguably broken) HTTP servers do not put the reason message in their
status line, which causes evhttp to consider the response invalid.
http://sourceforge.net/tracker/?func=detail&aid=2875077&group_id=50884&atid=461322
|
|
ae6ece02
|
2010-04-21T01:15:19
|
|
Clean up properly when adding a signal handler fails.
Previously, when a signation() or signal() call failed, we would free
the element we added to sh_old, but not actually clear the pointer.
This would leave a dangling pointer in sh_old that could cause a
crash later.
|
|
cb1a7223
|
2009-01-27T16:29:48
|
|
Fix memory leak when setting up priorities; reported by Alexander Drozdov
Backport from Libevent 2.0 commit 1c927b7
|
|
|
24589347
|
2010-03-26T14:22:47
|
|
Fix a free(NULL) in min_heap.h
Backport of 6f20492fa27f08
|
|
|
01ea0c5c
|
2010-02-03T16:54:18
|
|
make evhttp_send() safe against terminated connections, too
|
|
|
5c8b446e
|
2010-02-03T14:34:56
|
|
do not fail while sending on http connections the client closed.
when sending chunked requests via multiple calls to evhttp_send_reply_chunk,
the client may close the connection before the server is done sending. this
used to cause a crash.
we introduce a new function evhttp_request_get_connection() that allows the
server to determine if the request is still associated with a connection.
If it's not, evhttp_request_free() needs to be called explicitly or the user
can call evhttp_send_reply_end() which just frees the request, too.
|
|
|
20d706d0
|
2010-02-12T00:02:22
|
|
Distribute nmake makefile for 1.4
|
|
bce58d61
|
2010-02-03T23:27:40
|
|
Make Libevent 1.4.12 build on win32 with Unicode enabled.
This patch fixes calls to the win32 api to explicitly call the char* versions
of the functions. This fixes build failures when libevent is built with the
UNICODE define.
|
|
|
bd03d068
|
2010-01-14T18:22:00
|
|
Re-add event_siglcb; some old code _was_ still using it. :(
Such code really needs to migrate to use signal events instead.
This reverts commit 072ae5887e8064da500adbd2d24050bb96fbe75d.
|
|
4fd2dd9d
|
2010-01-11T18:52:54
|
|
Do not send an HTTP error when we've already closed or responded.
Previously, we'd issue an HTTP/1.1 400 Bad Request" response on every
connection close, event if sever sent response already.
This patch changes the behavior, so we only issue the response on
close when the connection state is not DISCONNECTED, and so we set
the state to DISCONNECTED when the connection closes.
Includes a regression test; fixes sourceforge bug 2909909.
|
|
|
8771d5b6
|
2010-01-07T02:25:48
|
|
Merge branch 'readln-backport' into patches-1.4
|
|
|
35df59e9
|
2010-01-04T01:01:54
|
|
Correct a debug message in evhttp_parse_request_line
The old message reported that evhttp_decode_uri had failed; in fact,
strdup had failed.
Found by Michael Lenaghan.
|
|
5f2e2507
|
2009-12-29T16:19:24
|
|
Adjusted fcntl() retval comparison on evutil_make_socket_nonblocking().
Apparently, a successful return value on F_SETFL is "anything but
-1".
|
|
81c26ba4
|
2009-12-28T16:00:05
|
|
Make evutil_make_socket_nonblocking() leave any other flags alone.
Fixes bug 2922121
|
|
|
ac0d213c
|
2009-12-18T23:50:42
|
|
Correct all versions in 1.4 branch
|
|
891765cd
|
2009-12-08T17:02:24
|
|
Update sample/signal-test.c to use newer APIs and not leak.
|
|
|
01f3775b
|
2009-11-21T01:11:49
|
|
Fix memory-leak of signal handler array with kqueue. [backport]
It turns out that kqueue_dealloc wasn't calling evsig_dealloc()
(because it doesn't use the main signal handler logic) so the sh_old
array was leaking.
This patch also introduces a fix in evsig_dealloc() where we set
the sh_old array to NULL when we free it, so that main/fork can pass.
|
|
5b10d008
|
2009-12-04T10:44:46
|
|
Fix a crash when reading badly formatted resolve.conf
Based on patch from Yasuoka Masahiko, backported to 1.4.
|
|
5713d5dd
|
2009-11-29T10:20:46
|
|
Valgrind fix: Clear struct kevent before checking for OSX bug.
William's original commit message:
Valgrind complains on startup because kq_init passes to kevent only
a partially initialized structure. The code doesn't expect kevent
to look at .fflags, .udata, or .data, I suppose, because it merely
tickles the kernel looking for an error response. But perhaps
that's unwarranted chuminess (notwithstanding that it's checking
for an OS X bug), and needless noise nonetheless.
|
|
|
c5454854
|
2009-11-23T13:22:15
|
|
Make the evbuffer_readln backport follow the current API
|
|
b04cc60f
|
2009-11-23T12:55:50
|
|
Backport evbuffer_readln().
This is necessary because it is not actually possible to use
evbuffer_readline() safely: it will treat "A\r\n" as 'A' EOL if it
reads it all at once, and as 'A' EOL EOL if there is a delay between
reading the \r and the \n.
Nicholas Marriott's comments on this patch:
Gilles is too busy so I've had a go at this, please see the diff
below. Rather than try to backport directly from 2.0 where the
evbuffer code is quite different, I've backported the _readln
function from when it was initially added in buffer.c r550. I can't
see any relevant bug fixes after this point so the function is
pretty much just copied in directly from that revision.
|
|
|
d014edb2
|
2009-11-20T12:20:49
|
|
Add a .gitignore file for the 1.4 branch.
|
|
|
06d26fc8
|
2009-11-18T04:11:22
|
|
call it 1.4.13-stable
svn:r1544
|
|
|
f76b599d
|
2009-11-16T19:21:25
|
|
Fix 1.4 compilation on msvc, and add a couple of minimal (not-quite-right) nmakefiles.
svn:r1534
|
|
|
25da0672
|
2009-11-16T18:46:10
|
|
Partial backport of event_rpcgen portability fixes.
svn:r1533
|
|
|
3487906f
|
2009-11-16T18:04:13
|
|
Add a missing include to make win32 compile.
svn:r1532
|
|
|
eb1fa9f7
|
2009-11-15T18:59:55
|
|
When running set[ug]id, don't check the environment.
Idea from OpenBSD, but made a bit more generic to handle uncivilized lands
that do not define issetugid.
svn:r1529
|
|
|
b4183c73
|
2009-11-09T19:55:40
|
|
Stop too many bytes for activequeues.
We were saying calloc(N,N*sizeof(struct event_list*)) when we should have
been saying calloc(N,sizeof(struct event_list*)). This wasted N*(N-1) words
of memory, where N was the number of priorities. This wouldn't be a big deal
for any sane number of priorities, but it's a bug nonetheless.
svn:r1526
|
|
|
e32d055b
|
2009-11-06T21:23:52
|
|
Backport GNU/kfreebsd kqueue fix
svn:r1515
|
|
|
6ce14fd0
|
2009-11-03T19:15:27
|
|
Rename compat/sys/_time.h to compat/sys/_libevent_time.h
On some systems (notably HPUX), there is already a
/usr/include/sys/_time.h, which our sys/_time.h shadows. Found and
diagnosed by Kathryn Hogg.
This is a quick fix for 1.4 only; for 2.0, I want to eliminate
compat/sys/_time.h entirely, and have util-internal subsume it.
svn:r1493
|
|
|
756c4cd3
|
2009-10-12T21:09:14
|
|
Declare struct timezone in evutil.h so that borken mingw versions do not complain
svn:r1442
|
|
|
c5a0f56c
|
2009-09-24T22:21:09
|
|
from trunk: Do not drop data from evbuffer when out of memory; reported by Jacek Masiulaniec
svn:r1437
|
|
|
03a9da17
|
2009-09-23T23:50:43
|
|
Fix compilation on Android: define fd_mask when it's missing.
svn:r1434
|
|
|
072ae588
|
2009-09-23T22:19:00
|
|
Remove unused event_gotsig code from 1.4 to appease some automated code checkers
svn:r1433
|
|
|
1bf18e5a
|
2009-09-11T18:55:12
|
|
Backport epoll memory fix for bug 2839240
svn:r1430
|
|
|
81765181
|
2009-09-11T18:55:06
|
|
Backport: do not believe negative result from FIONREAD.
svn:r1429
|
|
|
8c6282ec
|
2009-07-25T02:21:05
|
|
call it 1.4.12-stable
svn:r1376
|
|
|
9e652610
|
2009-07-23T00:24:28
|
|
A more hackish (but more backported) fix for the intptr vs uintptr fix from trunk.
svn:r1374
|
|
|
8155ed0c
|
2009-07-21T19:01:07
|
|
#ifdef out some callbacks in regress.c that windows doesn't use.
svn:r1371
|
|
|
4c364111
|
2009-07-21T18:57:38
|
|
Fix compilation of 1.4 branch on win32
svn:r1370
|
|
|
267472dc
|
2009-07-21T17:57:31
|
|
Do a quick-and-dirty hack to fix a gcc warning on 1.4 with a 64-bit arch
svn:r1368
|
|
|
9af9aba6
|
2009-07-21T17:57:25
|
|
Add -fno-strict-aliasing to let 1.4 build happily on GCC 4.4
svn:r1367
|
|
|
f5408f1d
|
2009-07-17T20:27:54
|
|
Checking for MS_WINDOWS rather than WIN32 is a Tor-ism. Fix that! {Backport}
svn:r1359
|
|
|
0ae18517
|
2009-07-17T19:00:22
|
|
Backport: Update event_tv when time jumps backwards, so that we only note each jump once. Not strictly needed in 1.4, but good for correctness.
svn:r1354
|
|
|
2bdf3077
|
2009-07-14T19:31:11
|
|
[Backport] Do not define _FORTIFY_SOURCE if the platform GCC already defined it for us.
svn:r1345
|
|
|
a763e741
|
2009-06-30T14:20:44
|
|
The truncated bit is in the 3rd byte of a dns reply, not the 4th.
svn:r1331
|
|
|
76339194
|
2009-06-24T22:39:12
|
|
Let evdns keep working when our IP changes. Fix by Christopher Davis; backported by nickm.
svn:r1328
|
|
|
b99254ac
|
2009-06-24T22:39:03
|
|
Activate fd events in a pseudorandom order on older backends.
New backends like poll and kqueue and so on add fds to the queue in
the order that they are triggered. But the select backend currently
activates low-numbered fds first, whereas the poll and win32 backends
currently favor whatever fds have been on for the longest. This is no
good for fairness.
svn:r1327
|
|
|
b0d88e68
|
2009-06-04T05:29:04
|
|
Fix epoll fencepost error. Patch most recently from Adam Langley, though I think I have seen others post this before.
svn:r1323
|
|
|
0ec290be
|
2009-05-25T18:15:44
|
|
When __GNUC__ is not defined, use __VA_ARGS__.
svn:r1312
|
|
|
9d0f6eb4
|
2009-05-22T18:32:09
|
|
Fix a potentially very annoying evdns bug that we found in Tor.
Generally speaking, it way better to event_assign() an event when you
allocate it than to assign it before every time you event_add it: if
it is already event_add()ed, the assign will mess it up so that it
doesn't _look_ added, and event_add() will insert a second copy.
Later, event_del() will only delete the second copy. Eventually, the
event_base will have a dangling pointer to freed memory. Ouch!
svn:r1308
|
|
|
6bd85f9d
|
2009-05-22T16:51:10
|
|
Fix compilation warning on win32 with verbose GCC warnings.
svn:r1305
|
|
|
499d0039
|
2009-05-22T15:09:43
|
|
Try to contain the failure when we are running without socketpair().
Some win32 systems (mostly those using Kaspersky, it would seem)
prevent us from faking socketpair(). This makes our signal
notification code just not work. Our response since 1.4 has been to
assert. For users who would rather work without signals than not work
at all, this has been a regression from 1.3e.
This patch makes adding signal events fail in this case; there's no
reason to kill the whole process.
svn:r1304
|
|
|
0f62db5e
|
2009-05-15T06:22:54
|
|
call it 1.4.11-stable
svn:r1290
|
|
|
9b45a4e7
|
2009-05-15T01:36:31
|
|
We were distributing the wrong event-config.h with our source distributions. Fix that.
svn:r1288
|
|
|
5d639d6c
|
2009-05-15T01:14:08
|
|
Fix win32 compilation warnings in 1.4.
svn:r1287
|
|
|
5e0563ba
|
2009-05-12T18:27:45
|
|
Fix dangling pointer in epoll after epoll_recalc().
This is based on patch 2790759 from Kevin Springborn. His comments on
sourceforge:
Problem:
The failure case is as follows: Event is added using epoll_add (a
direct pointer is stored in the user_data section), epoll_recalc is
called and the fds array is moved (invalidating the user_data
pointer stored in epoll). epoll_dispatch is called for the added
event and accesses evepoll based on the invalid pointer (set before
the fds array was relocated).
Solution:
Dispatch has access to the epollop structure, so given the fd we can
find the corresponding evepoll struct. Use data.fd instead of
data.ptr and store the fd corresponding to the event. This way when
epoll_recalc moves the fds array (and updates the fds array pointer
in epollop), the evepoll calculation in dispatch still finds the
valid evepoll struct.
svn:r1282
|
|
|
a276fa51
|
2009-05-12T18:27:34
|
|
Fix warnings in compilation of regress.c.
svn:r1281
|
|
|
23967f73
|
2009-04-24T16:08:30
|
|
Remove the limit on size of HTTP headers by removing static buffers
svn:r1240
|
|
|
749f4ce3
|
2009-04-23T00:02:19
|
|
[backport] Fix min_heap_erase when we remove an element from the middle of the heap.
Previously, we could lose the heap property when we removed an item
whose parent was greater than the last element in the heap. We would
replace the removed item with the last element, and consider shifting
it down, but we wouldn't consider shifting it up.
Patch from Marko Kreen.
svn:r1227
|
|
|
4d4a7d55
|
2009-04-17T00:37:21
|
|
call it 1.4.10
svn:r1178
|
|
|
1a518873
|
2009-04-12T23:00:52
|
|
from trunk: improve evhttp_parse_query documentation
svn:r1161
|
|
|
c1bd9385
|
2009-04-11T04:12:46
|
|
Fix parsing of queries where the encoded queries contained \r, \n or +
svn:r1155
|
|
|
64b3a571
|
2009-03-30T17:36:59
|
|
correct evbuffer_drain documentation from John Khvatov
svn:r1128
|
|
|
59513ee2
|
2009-03-12T17:45:13
|
|
from trunk: include Doxyfile in tar ball; from Jeff Garzik
svn:r1126
|
|
|
f06b29b9
|
2009-03-12T17:03:21
|
|
off-by-one error in epoll_recalc; reported by Victor Goya
svn:r1124
|
|
|
4cf8138b
|
2009-02-02T19:20:49
|
|
Make evdns_resolve_reverse args const.
svn:r1094
|
|
|
62d60546
|
2009-01-23T17:21:17
|
|
we dont use an m4 directory for 1.4 - remove it
svn:r1045
|
|
|
092baefa
|
2009-01-22T06:26:32
|
|
from trunk: fix signal processing for non-kqueue backends; when a signal callback delivers a signal; from Alexander Drozdov
svn:r1036
|
|
|
636e74a7
|
2009-01-22T02:49:57
|
|
from trunk: rename sys/signal.h to signal.h; configure m4 macro dir; this assist with compilation on Haiku
svn:r1034
|
|
|
7682ff80
|
2009-01-19T23:54:27
|
|
from trunk: rename the backend from "event ports" to "evport" - this will allow environment based disabling to work
svn:r1029
|