de36bc05f7add __likely branch prediction hints make inlining work in the face of gcc in non-gnu c99 mode
leitner
2010-11-05 03:07:47 +0000
faa5bad9fcfix io_receivefd so the incoming buffer is only 1 byte; io_passfd sends only one byte, so we might receive (and discard) other data if we try to read more (Stefan Bühler, from the lightttpd project)
leitner
2010-10-06 17:34:52 +0000
d2aae4864conly try to receive one byte to avoid discarding data
leitner
2010-10-06 17:33:25 +0000
928bfe5c09for installed header files, add comment saying where they came from
leitner
2010-06-03 22:46:59 +0000
3bed09a972get rid of warning
leitner
2010-01-19 02:44:06 +0000
3e16d08aa7update Makefile for bsd
leitner
2010-01-19 02:41:34 +0000
8e8788b8d5split make install
leitner
2010-01-19 02:40:40 +0000
acd4fdb68badd/fix man pages zero a buffer in io_passfd so valgrind does not complain when we pass it to a syscall (which did not use the uninitialized parts anyway)
leitner
2010-01-18 19:51:00 +0000
b4469f1ed8support SCTP in addition to TCP
leitner
2010-01-18 19:49:03 +0000
89b88f036dmake it compile again
leitner
2009-11-12 10:52:43 +0000
51202765fethe #define extern hack in io_fd.c also included time.h, fix that catch case in timeout loop where we have something to report on an fd; in that case do not report a timeout but let the other event get through first.
leitner
2009-04-25 19:09:27 +0000
67bfd53652check for socket() returning -1 before trying to set the socket non-blocking so we don't hide the "too many open files" error code
leitner
2008-09-10 09:51:26 +0000
97ac873038gcc 4.3.1 generates bad code for byte_copy, so I'm switching to a simpler version
leitner
2008-08-25 22:15:29 +0000
8583de1b55fix man page for openreadclose
leitner
2008-06-24 14:19:17 +0000
4fec25060bsmall man page updates and add a cdbget test
leitner
2007-04-30 05:08:28 +0000
b0d5046428fmt_ip6 compresses at best spot, not at first spot (Nikola Vladov)
leitner
2007-03-04 13:13:19 +0000
ff1805d310remove unnecessary casts in test/uudecode
leitner
2007-02-25 17:13:30 +0000
d3d6c828a9Turns out that shifting a negative signed int is undefined behavior in C, too. Use other mechanism in case gcc decides to abuse that, too.
leitner
2007-01-28 23:36:24 +0000
3e559e8d89remove a bogus gcc warning
leitner
2007-01-26 19:59:31 +0000
47d2e25ba3simplify some expressions
leitner
2007-01-26 19:56:41 +0000
7b52bfc982fix really pathological case where io_timeouted would never start over from the beginning because always new accept()ed connections came in and got newer, higher descriptors since the last io_timeouted loop. (Dirk Engling)
leitner
2007-01-24 18:28:39 +0000