Nick Mathewson
22845886c6
Merge remote branch 'origin/patches-2.0'
2011-01-31 16:45:23 -05:00
Nick Mathewson
3c8f4e758e
Add compile-time check for AF_UNSPEC==PF_UNSPEC
2011-01-31 16:44:06 -05:00
Nick Mathewson
6092f1265f
Fix http unit test on non-windows platforms without getaddrinfo
2011-01-31 16:37:27 -05:00
Nick Mathewson
cb92113979
Build on systems without AI_PASSIVE
2011-01-31 16:32:05 -05:00
Nick Mathewson
713c254d21
Try to build correctly on platforms with no IPv6 support
2011-01-24 18:55:10 -05:00
Nick Mathewson
9184563e49
Build correctly on platforms without sockaddr_storage
2011-01-24 18:29:44 -05:00
Nick Mathewson
453317b28c
Fall back to sscanf if we have no other way to implement strtoll
2011-01-24 18:22:32 -05:00
Nick Mathewson
da13d64486
Merge remote branch 'origin/patches-2.0'
2011-01-24 18:02:11 -05:00
Nick Mathewson
3267703658
Make --enable-gcc-warnings a no-op if not using gcc
2011-01-24 17:42:23 -05:00
Dave Hart
f1f85147f3
Allow use of --enable-silent-rules for quieter compilation with automake 1.11
2011-01-24 15:50:25 -05:00
Nick Mathewson
8b0afe96eb
Merge remote branch 'origin/patches-2.0'
2011-01-12 21:41:58 -05:00
Nick Mathewson
0c0ec0be2b
Correctly free selectop fields when select_resize fails in select_init
2011-01-12 20:28:47 -05:00
Nick Mathewson
83e805a415
Handle resize failures in the select backend better.
2011-01-07 13:18:09 -05:00
Jardel Weyrich
3f8d22a123
Use event_err() only if the failure is truly unrecoverable.
2011-01-07 13:03:32 -05:00
Jardel Weyrich
666b096691
Detect and handle more allocation failures.
2011-01-07 13:03:31 -05:00
Nick Mathewson
b647e0bc94
Merge remote branch 'kev009/21_ac_use_system_extensions'
2011-01-07 12:31:30 -05:00
Kevin Bowling
ded0a0907b
Add evconfig-private to remaining files
2011-01-07 00:41:37 -07:00
Kevin Bowling
fd7b5a81eb
Shell hack for weird mkdir -p commands
2011-01-07 00:37:05 -07:00
Kevin Bowling
c13e185914
Remove use and reference to event-private.h
2011-01-07 00:34:22 -07:00
Nick Mathewson
f1e9a7e7e4
Merge remote branch 'origin/patches-2.0'
2011-01-06 12:44:45 -05:00
Trond Norbye
0144886e7e
Check for POLLERR, POLLHUP and POLLNVAL for Solaris event ports
2011-01-06 12:43:53 -05:00
Kevin Bowling
f964b7215a
Prefer the ./configure evconfig-private.h in MinGW, just in case.
2011-01-03 15:44:26 -07:00
Kevin Bowling
ad03952edf
Backport libevent to vanilla Autoconf 2.59 (as used in RHEL5)
2011-01-03 15:09:21 -07:00
Kevin Bowling
b3651d1576
Add an evconfig-private.h stub for Win32.
2011-01-02 08:48:12 -07:00
Kevin Bowling
dc628c0304
Add an include guard
2011-01-02 08:47:46 -07:00
Kevin Bowling
0915ca0aa6
Include evconfig-private.h in internal files for great good.
2011-01-02 08:43:45 -07:00
Kevin Bowling
f6d66bc716
Fix a comment warning and add evconfig-private.h to .gitignore
2011-01-02 08:05:33 -07:00
Kevin Bowling
9b27b30720
Remove event-private.h and switch to evconfig-private.h
2011-01-02 08:04:12 -07:00
Kevin Bowling
868f88874f
Use a Configuration Header Template for evconfig-private.h
2011-01-02 07:51:23 -07:00
Kevin Bowling
edf62dddbc
Revert "evconfig-private.h with recusive configure subdirs"
...
This reverts commit ed53d5ef8053af08186ef80dfc1647cff769fac8.
2011-01-02 07:29:45 -07:00
Kevin Bowling
ed53d5ef80
evconfig-private.h with recusive configure subdirs
2011-01-02 06:35:05 -07:00
Kevin Bowling
ea8fa4cb23
Add AC_GNU_SOURCE to the fallback case.
2011-01-02 05:14:41 -07:00
Nick Mathewson
ac6904241b
Merge remote branch 'origin/patches-2.0'
2011-01-02 00:59:20 -05:00
Nick Mathewson
cb8059d2d3
Fix compilation on Windows with NDEBUG
...
Dongsheng Song reports that when building on windows with NDEBUG, you
run into an attempt to do EVUTIL_ASSERT(x) where x is a bitfield,
which turns into _EVUTIL_NIL_CONDITION(x), which takes sizeof(x),
which is illegal. This patch fixes _EVUTIL_NIL_CONDITION to work on
bitfields too.
2011-01-02 00:56:01 -05:00
Kevin Bowling
c51ef9307a
Eliminate a couple more manual internal _GNU_SOURCE defines
2010-12-22 23:24:01 -07:00
Kevin Bowling
3b265412b7
Remove internal usage of _GNU_SOURCE
2010-12-22 23:08:10 -07:00
Kevin Bowling
2c5566a9c7
Add event-private.h to noinst_HEADERS
2010-12-22 23:00:59 -07:00
Kevin Bowling
321b558793
Filter '# define' statements from autoconf and generate event-private.h
2010-12-22 22:53:03 -07:00
Kevin Bowling
1fa7dbe1e8
Add AC_USE_SYSTEM_EXTENSIONS to configure.in.
...
Requires follow on patches for correctness and robustness.
2010-12-22 22:11:26 -07:00
Nick Mathewson
41bb1cae79
Merge branch 'file_offset_v2'
2010-12-20 19:30:13 -05:00
Nick Mathewson
0aad01436a
Refactor addfile tests; add test for big files and for offsets
2010-12-20 19:25:15 -05:00
Nick Mathewson
3f405d2d4b
Add CreateFileMapping file_segment implementation for win32
2010-12-20 19:25:15 -05:00
Nick Mathewson
c2d9884a6a
Add support for mmaps with nonzero offset values. Needs testing.
2010-12-20 19:25:15 -05:00
Nick Mathewson
e72afae068
Add evbuffer_add_file_segment() so one fd can be used efficiently in more than one evbuffer_add_file at a time
2010-12-20 19:25:05 -05:00
Nick Mathewson
076680abe4
Merge branch 'epoll_table_v2'
2010-12-16 14:37:15 -05:00
Nick Mathewson
d4970d4edd
Reindent epoll_apply_one_change()
2010-12-16 14:37:05 -05:00
Nick Mathewson
2d55a190e1
Clean up error handling in epoll_apply_one_change() a little
...
The old code was more or less:
if (op == X && errno == FOO) {
...
} else if (op == Y && errno == BAR) {
...
}
but really we wanted to do a switch (op) to avoid needless checks
and branches.
This patch leaves the indentation a little weird so as to make it
easier to see what changed; the next patch will fix the indentation.
2010-12-16 14:34:14 -05:00
Nick Mathewson
8c83eb6948
Replace big chain of if/thens in epoll.c with a table lookup
...
This should save a bunch of branches by doing instead a lookup in a
nice static table.
To ensure correctness, the table is generated from a Python script,
included with this commit.
2010-12-16 14:17:49 -05:00
Nick Mathewson
b8b8aa560c
Use GetSystemTimeAsFileTime to implement gettimeofday on win32.
...
It is (in my benchmarks) way faster than _ftime, though the
conversion process is not so straightforward. In theory, it can
have a better granularity too, though in practice who knows what
you're getting.
2010-12-16 13:26:01 -05:00
Mike Smellie
04ba27ebf2
Use current event set rather than current pending change when deciding whether to no-op a del
...
This alters event_changelist_del to quash deletion of events that
didn't exist in the first place.
As far as I can see, the add,delete, dispatch case described in the
original comment will never happen. The recorded change is a single
operation, not a queue. This seems to leave actions to delete
events that never existed as the real targets for no-oping
2010-12-16 13:21:25 -05:00