Nick Mathewson
10c3450fd6
Make --enable-verbose-debug option match its help text
2011-08-31 11:58:40 -04:00
Nick Mathewson
4cbedae95f
Merge remote-tracking branch 'origin/patches-2.0'
...
Conflicts:
Makefile.am
WIN32-Code/event2/event-config.h
configure.in
2011-08-31 11:28:37 -04:00
Nick Mathewson
a718b6c31c
Increment version to 2.0.14-stable-dev
2011-08-31 11:25:11 -04:00
Nick Mathewson
ac9f2adb45
Credit new contributors for 2.0.14-stable
2011-08-31 00:23:20 -04:00
Nick Mathewson
da90a9be69
Pick a release date for the changelog
2011-08-30 22:35:51 -04:00
Nick Mathewson
85976ddfa4
Bump version to 2.0.14-stable
2011-08-30 22:28:02 -04:00
Zack Weinberg
148458e0a1
Use SIG_IGN instead of a do-nothing handler for signal events with kqueue
2011-08-30 15:33:02 -04:00
Zack Weinberg
46f1769d41
Fix OSX build: $(OPENSSL_INCS) needs to be after $(AM_CPPFLAGS).
2011-08-30 13:31:53 -04:00
Nick Mathewson
e9cccce6e1
Merge remote-tracking branch 'origin/patches-2.0'
2011-08-29 17:51:45 -04:00
Joachim Bauch
54f7e61ba3
clear read watermark on underlying bufferevent when creating filtering bev to fix potentially failing fragmented ssl handshakes
2011-08-29 17:50:34 -04:00
Nick Mathewson
6476d92d10
Checkpoint changelog entries for 2.0.14-stable
2011-08-29 13:40:03 -04:00
Nick Mathewson
904254f975
Merge remote-tracking branch 'origin/patches-2.0'
2011-08-28 14:51:36 -04:00
Nick Mathewson
e6af35d762
Correctly terminate IO on an async bufferevent on bufferevent_free
2011-08-28 14:39:12 -04:00
Nick Mathewson
0ff2c5a922
Have test-ratelim.c support IOCP
2011-08-28 14:39:12 -04:00
Nick Mathewson
a98da7bfc9
Make IOCP rate-limiting group support stricter and less surprising.
...
Previously, we wouldn't decrement read/write buckets because of IOCP
reads and writes until those reads and writes were complete. That's
not so bad on the per-connection front. But for group limits, the
old approach makes us launch a huge amount of reads and writes
whenever the group limit becomes positive, and then decrement the
limit to a hugely negative number as they complete.
With this patch, we decrement our read buckets whenever we launch an
IOCP read or write, based on the maximum that tried to read or
write. Later, when the operations finish, we re-increment the
bucket based on the portion of the request that couldn't finish.
2011-08-28 14:39:11 -04:00
Nick Mathewson
c75341b077
Support negative arguments to _bufferevent_decrement_(read/write)_buckets()
2011-08-28 14:39:11 -04:00
Harlan Stenn
6d2613becc
We need AM_CPPFLAGS when compiling bufferevent_openssl.c
2011-08-28 14:31:37 -04:00
Harlan Stenn
1c63860758
Use pkg-config (if available) to handle OpenSSL.
2011-08-28 14:31:28 -04:00
Harlan Stenn
674dc3d34e
Provide the autoconf m4 macros for the new OpenSSL via pkg-config stuff.
2011-08-28 14:31:28 -04:00
Nick Mathewson
b260065ab6
Merge remote-tracking branch 'origin/patches-2.0'
2011-08-28 14:06:03 -04:00
Nick Mathewson
2f51dc0311
Cleanup on 7c11e51e1ab: fix strtol usage
2011-08-28 14:03:10 -04:00
Nick Mathewson
6056d6e0df
Cleanup on 7c11e51e1ab: restore c90 declaration compliance
2011-08-28 14:02:40 -04:00
Harlan Stenn
7c11e51e1a
Clean up some problems identified by Coverity.
2011-08-28 13:44:33 -04:00
Nick Mathewson
2cbe115cbc
Merge remote-tracking branch 'origin/patches-2.0'
2011-08-24 16:17:56 -04:00
Nick Mathewson
5b18f13048
Make rate limiting work with common_timeout logic
2011-08-24 16:17:05 -04:00
Nick Mathewson
0d10804465
Merge remote-tracking branch 'origin/patches-2.0'
2011-08-18 15:18:14 -04:00
Nick Mathewson
6736852eaa
Merge branch '20_iocp_fixes' into patches-2.0
2011-08-18 15:09:44 -04:00
Nick Mathewson
6acfbdd83a
Make overlapped reads result in evbuffer callbacks getting invoked
2011-08-18 13:15:46 -04:00
Nick Mathewson
495c227f59
IOCP: don't launch reads or writes on an unconnected socket
2011-08-18 11:41:55 -04:00
Nick Mathewson
ec670c15a7
Merge remote-tracking branch 'origin/patches-2.0'
...
Conflict in buffer.c: the new file-segment logic conflicted with the
solaris sendfile fix.
2011-08-17 22:08:06 -04:00
Dave Hart
04656ea85a
Try to fix 'make distcheck' errors when building out-of-tree
2011-08-17 22:04:38 -04:00
Nick Mathewson
3c63edd1f7
Make the priority inversion code use gettime(), not evutil_gettimeofday()
...
Since we're computing the time after each callback, we might as well
update the time cache (if we're using it) and use monotonic time (if
we've got that).
2011-08-17 22:04:04 -04:00
Nick Mathewson
a37a0c0e35
Make max_dispatch_interval able to apply only to low-priority events
...
Suggested by Alexander Drozdov
2011-08-17 22:03:57 -04:00
Alexander Drozdov
a9866aa8c1
Optimization in event_process_active(): ignore maxcb & endtime for highest priority events.
2011-08-17 22:03:57 -04:00
Nick Mathewson
9fa56bdf1c
Make sure max_dispatch_callbacks is never negative
...
Suggested by Alexander Drozdov.
2011-08-17 22:03:57 -04:00
Nick Mathewson
fd4de1e7fe
Add event_config function to limit time/callbacks between calls to dispatch
2011-08-17 22:03:57 -04:00
Nick Mathewson
e067d0e3ac
Merge branch '20_low_ratelim' into patches-2.0
2011-08-17 21:47:19 -04:00
Michael Herf
643922e907
Solaris sendfile: correctly detect amount of data sent
...
Original message:
Solaris sendfile seems to fail when sending moderately large (<1GB)
files. Not a 32/64 problem, but a buffer problem.
Anyone else ever try this? It is definitely broken in http-server.c.
It seems to be broken in the following way:
When sendfile sends partial data (EAGAIN, would block), "res" is
always -1, rather than the amount sent.
Here's a patch that reads from the "offset" pointer instead to
discover what was sent. This seems to work:
2011-08-15 13:39:10 -04:00
Nick Mathewson
2fdb8e2204
Merge remote-tracking branch 'ellzey/debug_timer_add_usec'
2011-08-15 11:29:39 -04:00
Nick Mathewson
d7451b0523
Merge remote-tracking branch 'origin/patches-2.0'
2011-08-15 11:14:57 -04:00
Leonid Evdokimov
aff6ba15a1
Fix request_finished memory leak with debugging turned on.
2011-08-15 11:12:57 -04:00
Leonid Evdokimov
9b724b288a
Fix evsig_dealloc memory leak with debugging turned on.
2011-08-15 11:12:52 -04:00
Leonid Evdokimov
50be5a14b5
Another docstring fix.
2011-08-15 11:11:10 -04:00
Mark Ellzey
67275433ec
more event dbg updates
2011-08-11 16:53:01 -05:00
Nick Mathewson
6d5440e80e
Fix handling of group rate limits under 64 bytes of burst
...
The "min_share" logic, which was designed to prevent piles of
extremely small writes when running up against a group rate limit,
could lead to confusing behavior if you ever set a min_share less
than your burst rate. If that happened, then as soon as your group
rate limit was exhausted, you'd stop reading/writing, and never
start again, since the amount readable/writeable would never
actually hit min_share.
We now cap min_share at the rate per tick.
Found by George Kadianakis
2011-08-11 15:15:17 -04:00
Mark Ellzey
4b7d298415
added timeout debug logs to include event ptr.
2011-08-11 13:25:24 -05:00
Nick Mathewson
14ec20a7a3
Merge remote-tracking branch 'ellzey/debug_timer_add_usec'
2011-08-11 14:00:01 -04:00
Nick Mathewson
d87a3f71c6
Merge remote-tracking branch 'ellzey/event_debug_flag'
2011-08-11 13:59:02 -04:00
Mark Ellzey
3baab0dce9
Added usec debug in another area for debug
2011-08-11 12:06:54 -05:00
Mark Ellzey
ac43ce0450
Debug addition for printing usec on TIMEOUT debugging.
2011-08-11 11:56:26 -05:00
