leitner
|
12caccfd48
|
don't shrink in array_allocate
|
20 years ago |
|
leitner
|
e1e80badf0
|
add some debug stuff
|
20 years ago |
|
leitner
|
46dcfbdce1
|
Uh, it's #warning and not #warn. *blush*
|
20 years ago |
|
leitner
|
289ca04238
|
Work around OpenBSD. Again.
|
20 years ago |
|
leitner
|
f0e9a2afc9
|
work around broken Solaris sys/socket.h
|
20 years ago |
|
leitner
|
c169ae9b4a
|
IRIX compatibility (yuck!)
|
20 years ago |
|
leitner
|
9405f1271b
|
remove gcc4-only warning option
|
20 years ago |
|
leitner
|
c874604f12
|
gcc 4 cleanups (mostly unsigned char* vs char*)
|
20 years ago |
|
leitner
|
7cd919d333
|
silence two gcc warnings
|
20 years ago |
|
leitner
|
11da04b651
|
fix descriptor leak in iob_addfile_close if the range was 0 (oops)
|
20 years ago |
|
leitner
|
1205a3d38f
|
byte_zero the allocated io_batch in iob_new
|
20 years ago |
|
leitner
|
882885fdff
|
remove cruft
|
20 years ago |
|
leitner
|
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 |
|
leitner
|
e73c3e85f1
|
document error case
|
20 years ago |
|
leitner
|
bbf4c614bf
|
fix iob_write to handle failure properly
document that the iob_write callback should limit itself
|
20 years ago |
|
leitner
|
f4c8a0c7c2
|
fix openbsd problem?
|
20 years ago |
|
leitner
|
fd4344f3c0
|
how could this ever have worked?
|
20 years ago |
|
leitner
|
89c12ba545
|
typo, found by Tino Reichardt
|
20 years ago |
|
leitner
|
b617d63d1e
|
update dependencies
add errmsg_info and msg
|
20 years ago |
|
leitner
|
bffb1dab90
|
mention prototype of io_write_callback
|
20 years ago |
|
leitner
|
86cfc33dab
|
add iob_write (send io batch through callback, e.g. for SSL)
|
20 years ago |
|
leitner
|
b48ce77792
|
add iob_write (to write an io batch through a callback)
|
20 years ago |
|
leitner
|
e95599085a
|
use madvise MADV_PREFETCH to implement iob_prefetch
|
21 years ago |
|
leitner
|
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 |
|
leitner
|
523f6f4765
|
cosmetics: use write instead of writev if there is only one buffer to
write (looks nicer in strace)
|
21 years ago |
|
leitner
|
72382af11a
|
Solaris compatibility for io_passfd and io_receivefd (untested, but at
least it compiles)
|
21 years ago |
|
leitner
|
1e871b1c32
|
actually return 0 on success in io_passfd
|
21 years ago |
|
leitner
|
696c5f872c
|
add io_readwritefile, io_appendfile
|
21 years ago |
|
leitner
|
11f0616cf3
|
work around broken Linux sendfile API (offset 64-bit but count 32-bit)
|
21 years ago |
|
leitner
|
768a46e304
|
argh more bsd
|
21 years ago |
|
leitner
|
a23757a317
|
bsd compat
|
21 years ago |
|
leitner
|
f26451c086
|
slight optimization for sending 0 bytes from a file
(also fixes a FreeBSD-current sendfile breakage)
|
21 years ago |
|
leitner
|
5e2141e9f6
|
fix bug in mmap code path of io_sendfile (David Leadbeater)
|
21 years ago |
|
leitner
|
d8ce4dc2e5
|
fix signal race (Scott Lamb)
|
21 years ago |
|
leitner
|
076a8f4628
|
explicitly define extern globals
|
21 years ago |
|
leitner
|
78d181cb3e
|
document return values correctly
|
21 years ago |
|
leitner
|
6fb76c3cd9
|
<io.h> -> <iob.h>
|
21 years ago |
|
leitner
|
4aecd26995
|
add more man pages
|
21 years ago |
|
leitner
|
24227e0e01
|
add sigpipe helper
|
21 years ago |
|
leitner
|
1d9075071f
|
add some man pages
|
21 years ago |
|
leitner
|
b27aa1a9d8
|
better error handling
|
21 years ago |
|
leitner
|
70e112b892
|
add fd passing functions
|
21 years ago |
|
leitner
|
745c4affc5
|
add io_socketpair
|
21 years ago |
|
leitner
|
3996ade63e
|
add iob_addfile_close (iob_reset will close the files)
|
21 years ago |
|
leitner
|
adcb7aee21
|
handle POLLNVAL (can't happen errors)
|
21 years ago |
|
leitner
|
8454eb7e4c
|
speculative pre-emptive freebsd 4 compatibility ;)
|
21 years ago |
|
leitner
|
069aedc871
|
try to fix freebsd 4 compile error
|
21 years ago |
|
leitner
|
4eed466fbc
|
writev fix
|
21 years ago |
|
leitner
|
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 |
|
leitner
|
c18a0dbf62
|
actually check in AIX sendfile code
|
21 years ago |
|
leitner
|
cae1a561fd
|
integrate Linux sendfile into trysendfile.c infrastructure
first stab at AIX 5 sendfile support
|
21 years ago |
|
leitner
|
7bcba8b2d1
|
still had bus error :(
|
21 years ago |
|
leitner
|
91060a1772
|
still bus error
|
21 years ago |
|
leitner
|
6d35239e2c
|
bus error?!
|
21 years ago |
|
leitner
|
e92d4ca5f5
|
prefetching leaked memory
|
21 years ago |
|
leitner
|
f21e1c57a3
|
macos x support
|
21 years ago |
|
leitner
|
8f188cc95b
|
macos x support
|
21 years ago |
|
leitner
|
7ddce27f49
|
Support Solaris 9 sendfile
|
21 years ago |
|
leitner
|
4740036c6f
|
#define _LARGEFILE64_SOURCE
|
21 years ago |
|
leitner
|
f484ecdc5f
|
support HP-UX sendfile (thanks Rolf Eike Beer)
|
21 years ago |
|
leitner
|
8ce9e6c016
|
sendfile
|
21 years ago |
|
leitner
|
2a516c901a
|
recent change corrected sendfile parameters but not internal usage
|
21 years ago |
|
leitner
|
fa476a9200
|
add experimental prefetching support
|
21 years ago |
|
leitner
|
f52aa87441
|
more alloca fixes
|
21 years ago |
|
leitner
|
876738bd8e
|
make cast nicer
|
21 years ago |
|
leitner
|
03461c6eb1
|
more defensive programming for SIGIO (*groan*)
|
21 years ago |
|
leitner
|
3cac9c2b80
|
make sure the mmap based io_sendfile actually works (and terminates)
|
21 years ago |
|
leitner
|
9a0e23f420
|
oops, left debug #undef in
|
21 years ago |
|
leitner
|
a17f7c30d2
|
use mmap on BSD without sendfile
|
21 years ago |
|
sanjiyan
|
a58d400d5c
|
fixing the poll call in io_waituntil2
|
21 years ago |
|
leitner
|
0a38894c43
|
*duh* one extra zero
|
21 years ago |
|
sanjiyan
|
29cfd8df4e
|
reset timer correct.
|
21 years ago |
|
sanjiyan
|
90bde5e569
|
reset the timer in the correct way.
|
21 years ago |
|
leitner
|
6622288c7c
|
make BSD sendfile actually work
|
21 years ago |
|
leitner
|
ba849a2720
|
duh!!! Typo prevented iob_send from using FreeBSD sendfile
|
21 years ago |
|
leitner
|
a4cf7a8a8b
|
fix Solaris /dev/poll
|
21 years ago |
|
leitner
|
65adb024f5
|
bug fixing time
|
21 years ago |
|
leitner
|
82a4ae868d
|
broken lose32 port using mingw (don't use!!!)
initial Solaris /dev/poll port
|
21 years ago |
|
leitner
|
b81e8bcd27
|
typo
|
21 years ago |
|
leitner
|
17f3955f84
|
handle BSD errors
|
21 years ago |
|
leitner
|
361872ffa4
|
handle errors properly
|
21 years ago |
|
leitner
|
96be549818
|
handle errors signalled by epoll
|
21 years ago |
|
leitner
|
0335358787
|
add io_finishandshutdown
|
21 years ago |
|
leitner
|
0ba7b25005
|
make iob_reset actually free the buffers
|
21 years ago |
|
leitner
|
543987da58
|
remove superfluous #includes
make it possibly to specify that a buffer should be freed
|
21 years ago |
|
leitner
|
27ffe564a7
|
set flag only if setting to non-blocking worked
|
21 years ago |
|
leitner
|
fcd6082250
|
did not compile on BSD after SIGIO fix
|
21 years ago |
|
leitner
|
209fef467e
|
Big Duh! _SIGIO and SIGIO look too close
|
21 years ago |
|
leitner
|
ec060d6dee
|
wrong fix
|
21 years ago |
|
leitner
|
07a205405b
|
Add DEBUG=1 to GNUmakefile
the epoll guard broke SIGIO
|
21 years ago |
|
leitner
|
963fa817de
|
guard against one descriptor being flagged twice
|
21 years ago |
|
leitner
|
b01e72af69
|
quick fix to non-block issue
|
21 years ago |
|
leitner
|
eaa4dea67b
|
correctly initialize records before reusing them in io_fd
do something about unexpected epoll events
|
22 years ago |
|
leitner
|
29ae01d54f
|
glibc compatibility ;)
|
22 years ago |
|
leitner
|
2e1da4790c
|
alloca may be defined in stdlib.h
|
22 years ago |
|
leitner
|
d00a672dad
|
alloca warning removal
|
22 years ago |
|
leitner
|
f6b5de0c93
|
forgot to include <unistd.h>
|
22 years ago |
|
leitner
|
b97909c538
|
remove warning
|
22 years ago |
|
leitner
|
0c64df2331
|
make BSD sendfile optional (NetBSD doesn't have it)
|
22 years ago |
|
leitner
|
97043d279f
|
make io_waituntil actually take an "until", not a "how long"
fix io_timeouted to ignore fds where no timeout has been set
|
22 years ago |