953eb639b2
on Linux, save a few syscalls by using sendto/sendmsg with MSG_MORE
...
instead of write/writev + setsockopt TCP_CORK
10 years ago
4404ab35f3
revert to level triggering for epoll (forgot io_wantread)
10 years ago
967e3ce019
revert edge triggering epoll; it had reliability and fairness issues and
...
was also not actually faster
10 years ago
39e7ee90bb
introduce io_eagain_read and io_eagain_write (discontinue using io_eagain plz)
11 years ago
d9cbb3940c
it turns out you don't need -1/EGAIN to know edge triggering told you
...
there is no more data, it's enough to read/write less than you asked for
11 years ago
5baf943f0a
finish epoll move to edge triggering
11 years ago
a544abc39c
switch io_fds from array to newly implemented (hopefully thread-safe) iarray
...
switch epoll from level triggering to edge triggering
11 years ago
ff3a3410ec
hopefully this fixes the looping gatling processes
11 years ago
e175800a8c
add io_fd_canwrite (like io_fd but assume the fd is writable)
...
save a few syscalls here and there
13 years ago
6e6fc8b85d
fix endless loop bug on freebsd
13 years ago
c2a2a15c12
document the return value expected from the callback given to iob_write
...
remove unused #include in iob_reset
if iob_addfile_close fails, it now closes the fd
if iob_addbuf_munmap fails, it now munmaps the buf
if iob_addbuf_free fails, it now frees the buf
some win32 cross-compile fixes for iarray
13 years ago
662cf55fc9
accept -3 as error return from the callback, too
13 years ago
b0886bc324
the iob calls return int, not void
13 years ago
d2aae4864c
only try to receive one byte to avoid discarding data
14 years ago
3bed09a972
get rid of warning
15 years ago
acd4fdb68b
add/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)
15 years ago
1fcfcd268a
hups
16 years ago
51202765fe
the #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.
16 years ago
fd47173110
remove four warnings
16 years ago
a8a5e9a3d5
some valgrind de-noising
...
fix subtle typo in sub_of (David Sirovsky)
17 years ago
68d88f60b9
munmap cleanup fix (Erdgeist)
17 years ago
8b17690318
use callback based infrastructure to reset iob's
17 years ago
527efc70b6
add iob_addbuf_munmap
17 years ago
1b88982d19
off-by-one
18 years ago
7b52bfc982
fix 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)
18 years ago
8eab2b5aaf
make io_timeout clearer
18 years ago
f046d1cdc1
add iob_bytesleft
...
don't leak memory in test/httpd.c
18 years ago
83c1186870
fix for NetBSD 3.0 (they broke their sys/types.h)
...
remove support for ip6.int reverse lookup in DNS routines (the domain is
no longer delegated)
19 years ago
d361d81c64
make socket_(tc|ud)p[46] actually return non-blocking sockets as
...
documented (Richard Lyons)
19 years ago
d468ea9eb7
add io_block
19 years ago
c16d9880dc
remove printf
19 years ago
c1c50c7dbd
further Windoze support (test/io5.c works, gatling still doesn't)
...
This is just to get gatling to work, I may remove it again after
that.
19 years ago
f65398ff2e
io_close resets the cookie
20 years ago
9512cee614
add -D_REENTRANT to CFLAGS so libowfat can be used in multi-threaded
...
programs
20 years ago
b1e8f341ba
no debug
20 years ago
12caccfd48
don't shrink in array_allocate
20 years ago
e1e80badf0
add some debug stuff
20 years ago
46dcfbdce1
Uh, it's #warning and not #warn. *blush*
20 years ago
289ca04238
Work around OpenBSD. Again.
20 years ago
f0e9a2afc9
work around broken Solaris sys/socket.h
20 years ago
c169ae9b4a
IRIX compatibility (yuck!)
20 years ago
9405f1271b
remove gcc4-only warning option
20 years ago
c874604f12
gcc 4 cleanups (mostly unsigned char* vs char*)
20 years ago
7cd919d333
silence two gcc warnings
20 years ago
11da04b651
fix descriptor leak in iob_addfile_close if the range was 0 (oops)
20 years ago
1205a3d38f
byte_zero the allocated io_batch in iob_new
20 years ago
882885fdff
remove cruft
20 years ago
9eb09b5bfe
add iob_free and man page for iob_reset.
...
check in some windoze compat crap (still does not compile through for
windoze)
20 years ago
e73c3e85f1
document error case
20 years ago
bbf4c614bf
fix iob_write to handle failure properly
...
document that the iob_write callback should limit itself
20 years ago
f4c8a0c7c2
fix openbsd problem?
20 years ago
fd4344f3c0
how could this ever have worked?
20 years ago
89c12ba545
typo, found by Tino Reichardt
20 years ago
b617d63d1e
update dependencies
...
add errmsg_info and msg
20 years ago
bffb1dab90
mention prototype of io_write_callback
20 years ago
86cfc33dab
add iob_write (send io batch through callback, e.g. for SSL)
20 years ago
b48ce77792
add iob_write (to write an io batch through a callback)
20 years ago
e95599085a
use madvise MADV_PREFETCH to implement iob_prefetch
20 years ago
f4108178f2
fix case where iob_send sends all from io batch, but then new data is
...
added and iob_send is called again
21 years ago
523f6f4765
cosmetics: use write instead of writev if there is only one buffer to
...
write (looks nicer in strace)
21 years ago
72382af11a
Solaris compatibility for io_passfd and io_receivefd (untested, but at
...
least it compiles)
21 years ago
1e871b1c32
actually return 0 on success in io_passfd
21 years ago
696c5f872c
add io_readwritefile, io_appendfile
21 years ago
11f0616cf3
work around broken Linux sendfile API (offset 64-bit but count 32-bit)
21 years ago
768a46e304
argh more bsd
21 years ago
a23757a317
bsd compat
21 years ago
f26451c086
slight optimization for sending 0 bytes from a file
...
(also fixes a FreeBSD-current sendfile breakage)
21 years ago
5e2141e9f6
fix bug in mmap code path of io_sendfile (David Leadbeater)
21 years ago
d8ce4dc2e5
fix signal race (Scott Lamb)
21 years ago
076a8f4628
explicitly define extern globals
21 years ago
78d181cb3e
document return values correctly
21 years ago
6fb76c3cd9
<io.h> -> <iob.h>
21 years ago
4aecd26995
add more man pages
21 years ago
24227e0e01
add sigpipe helper
21 years ago
1d9075071f
add some man pages
21 years ago
b27aa1a9d8
better error handling
21 years ago
70e112b892
add fd passing functions
21 years ago
745c4affc5
add io_socketpair
21 years ago
3996ade63e
add iob_addfile_close (iob_reset will close the files)
21 years ago
adcb7aee21
handle POLLNVAL (can't happen errors)
21 years ago
8454eb7e4c
speculative pre-emptive freebsd 4 compatibility ;)
21 years ago
069aedc871
try to fix freebsd 4 compile error
21 years ago
4eed466fbc
writev fix
21 years ago
177e18f415
the mmap+write version of io_sendfile sent too much at the end and/or
...
went into an infinite loop sometimes.
21 years ago
c18a0dbf62
actually check in AIX sendfile code
21 years ago
cae1a561fd
integrate Linux sendfile into trysendfile.c infrastructure
...
first stab at AIX 5 sendfile support
21 years ago
7bcba8b2d1
still had bus error :(
21 years ago
91060a1772
still bus error
21 years ago
6d35239e2c
bus error?!
21 years ago
e92d4ca5f5
prefetching leaked memory
21 years ago
f21e1c57a3
macos x support
21 years ago
8f188cc95b
macos x support
21 years ago
7ddce27f49
Support Solaris 9 sendfile
21 years ago
4740036c6f
#define _LARGEFILE64_SOURCE
21 years ago
f484ecdc5f
support HP-UX sendfile (thanks Rolf Eike Beer)
21 years ago
8ce9e6c016
sendfile
21 years ago
2a516c901a
recent change corrected sendfile parameters but not internal usage
21 years ago
fa476a9200
add experimental prefetching support
21 years ago
f52aa87441
more alloca fixes
21 years ago
876738bd8e
make cast nicer
21 years ago
leitner